diff --git a/.gitattributes b/.gitattributes
index a6344aac8c09253b3b630fb776ae94478aa0275b..f2c00929a328c5db54fed585a8a8205a3b1e2cc6 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+resources/code_en.gif filter=lfs diff=lfs merge=lfs -text
+resources/web-demo.gif filter=lfs diff=lfs merge=lfs -text
+resources/web-demo2.gif filter=lfs diff=lfs merge=lfs -text
diff --git a/.github/ISSUE_TEMPLATE/bug_report.yaml b/.github/ISSUE_TEMPLATE/bug_report.yaml
new file mode 100644
index 0000000000000000000000000000000000000000..580d03eedaa576fdc4895bb091b31a75533f829f
--- /dev/null
+++ b/.github/ISSUE_TEMPLATE/bug_report.yaml
@@ -0,0 +1,77 @@
+name: "\U0001F41B Bug Report"
+description: Submit a bug report to help us improve ChatGLM3 / 提交一个 Bug 问题报告来帮助我们改进 ChatGLM3
+body:
+  - type: textarea
+    id: system-info
+    attributes:
+      label: System Info / 系統信息
+      description: Your operating environment / 您的运行环境信息
+      placeholder: Includes Cuda version, Transformers version, Python version, operating system, hardware information (if you suspect a hardware problem)... / 包括Cuda版本，Transformers版本，Python版本，操作系统，硬件信息(如果您怀疑是硬件方面的问题)...
+    validations:
+      required: true
+
+  - type: textarea
+    id: who-can-help
+    attributes:
+      label: Who can help? / 谁可以帮助到您？
+      description: |
+        Your issue will be replied to more quickly if you can figure out the right person to tag with @
+        All issues are read by one of the maintainers, so if you don't know who to tag, just leave this blank and our maintainer will ping the right person.
+    
+        Please tag fewer than 3 people.
+        
+        如果您能找到合适的标签 @，您的问题会更快得到回复。
+        所有问题都会由我们的维护者阅读，如果您不知道该标记谁，只需留空，我们的维护人员会找到合适的开发组成员来解决问题。
+        
+        标记的人数应该不超过 3 个人。
+
+        Related demo leader / 相关demo负责人 :
+        - finetune_demo: @Btlmd
+        - langchain_demo: @yincf
+        - composite_demo: @abmfy
+        
+        If it's not a bug in these three subsections, you may not specify the helper. Our maintainer will find the right person in the development group to solve the problem.
+        
+        如果不是这三个子版块的bug，您可以不指明帮助者，我们的维护人员会找到合适的开发组成员来解决问题。
+
+      placeholder: "@Username ..."
+
+  - type: checkboxes
+    id: information-scripts-examples
+    attributes:
+      label: Information / 问题信息
+      description: 'The problem arises when using: / 问题出现在'
+      options:
+        - label: "The official example scripts / 官方的示例脚本"
+        - label: "My own modified scripts / 我自己修改的脚本和任务"
+
+  - type: textarea
+    id: reproduction
+    validations:
+      required: true
+    attributes:
+      label: Reproduction / 复现过程
+      description: |
+        Please provide a code example that reproduces the problem you encountered, preferably with a minimal reproduction unit.
+        If you have code snippets, error messages, stack traces, please provide them here as well.
+        Please format your code correctly using code tags. See https://help.github.com/en/github/writing-on-github/creating-and-highlighting-code-blocks#syntax-highlighting
+        Do not use screenshots, as they are difficult to read and (more importantly) do not allow others to copy and paste your code.
+        
+        请提供能重现您遇到的问题的代码示例,最好是最小复现单元。
+        如果您有代码片段、错误信息、堆栈跟踪，也请在此提供。
+        请使用代码标签正确格式化您的代码。请参见 https://help.github.com/en/github/writing-on-github/creating-and-highlighting-code-blocks#syntax-highlighting
+        请勿使用截图，因为截图难以阅读，而且（更重要的是）不允许他人复制粘贴您的代码。
+      placeholder: |
+        Steps to reproduce the behavior/复现Bug的步骤:
+          
+          1.
+          2.
+          3.
+
+  - type: textarea
+    id: expected-behavior
+    validations:
+      required: true
+    attributes:
+      label: Expected behavior / 期待表现
+      description: "A clear and concise description of what you would expect to happen. /简单描述您期望发生的事情。"
\ No newline at end of file
diff --git a/.github/ISSUE_TEMPLATE/feature-request.yaml b/.github/ISSUE_TEMPLATE/feature-request.yaml
new file mode 100644
index 0000000000000000000000000000000000000000..8a9bf880909b4b4c3c9f344182fe28b4d28e3992
--- /dev/null
+++ b/.github/ISSUE_TEMPLATE/feature-request.yaml
@@ -0,0 +1,34 @@
+name: "\U0001F680 Feature request"
+description: Submit a request for a new ChatGLM3 feature / 提交一个新的 ChatGLM3 的功能建议
+labels: [ "feature" ]
+body:
+  - type: textarea
+    id: feature-request
+    validations:
+      required: true
+    attributes:
+      label: Feature request  / 功能建议
+      description: |
+        A brief description of the functional proposal. Links to corresponding papers and code are desirable.
+        对功能建议的简述。最好提供对应的论文和代码链接
+
+  - type: textarea
+    id: motivation
+    validations:
+      required: true
+    attributes:
+      label: Motivation / 动机
+      description: |
+        Your motivation for making the suggestion. If that motivation is related to another GitHub issue, link to it here.
+        您提出建议的动机。如果该动机与另一个 GitHub 问题有关，请在此处提供对应的链接。
+
+  - type: textarea
+    id: contribution
+    validations:
+      required: true
+    attributes:
+      label: Your contribution / 您的贡献
+      description: |
+        
+        Your PR link or any other link you can help with.
+        您的PR链接或者其他您能提供帮助的链接。
\ No newline at end of file
diff --git a/.github/PULL_REQUEST_TEMPLATE/pr_template.md b/.github/PULL_REQUEST_TEMPLATE/pr_template.md
new file mode 100644
index 0000000000000000000000000000000000000000..1830e407063f8fdc563adf4041f3e191510e7257
--- /dev/null
+++ b/.github/PULL_REQUEST_TEMPLATE/pr_template.md
@@ -0,0 +1,34 @@
+#  Raise valuable PR / 提出有价值的PR
+
+## Caution/ 注意事项:
+Users should keep the following points in mind when submitting PRs:
+
+1. The proposed PR should be about this project. 
+2. the proposed PR should be relevant, if there are multiple ideas and optimizations, they should be assigned to different PRs.
+
+用户在提交PR时候应该注意以下几点:
+
+1. 提出的PR应该是关于本项目的。
+2. 提出的PR应该具有针对性，如果具有多个不同的想法和优化方案，应该分配到不同的PR中。
+
+## 不应该提出的PR / PRs that should not be proposed
+
+If a developer proposes a PR about any of the following, it may be closed or Rejected.
+
+1. those that don't describe improvement options.
+2. multiple issues of different types combined in one PR.
+3. The proposed PR is highly duplicative of already existing PRs.
+
+如果开发者提出关于以下方面的PR，则可能会被直接关闭或拒绝通过。
+
+1. 没有说明改进方案的。
+2. 多个不同类型的问题合并在一个PR中的。
+3. 提出的PR与已经存在的PR高度重复的。
+
+
+# 检查您的PR
+- [ ] Have you read the Contributor Guidelines, Pull Request section? / 您是否阅读了贡献者指南、Pull Request 部分？
+- [ ] Has this been discussed/approved via a Github issue or forum? If so, add a link. / 是否通过 Github 问题或论坛讨论/批准过？如果是，请添加链接。
+- [ ] Did you make sure you updated the documentation with your changes? Here are the Documentation Guidelines, and here are the Documentation Formatting Tips. /您是否确保根据您的更改更新了文档？这里是文档指南，这里是文档格式化技巧。
+- [ ] Did you write new required tests? / 您是否编写了新的必要测试？
+- [ ]  Are your PRs for only one issue / 您的PR是否仅针对一个问题
\ No newline at end of file
diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000000000000000000000000000000000000..ab2af567c680ecb2b775c27c88bc669ceda2dd11
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,10 @@
+__pycache__
+
+# finetune_demo: generated & downloaded files
+finetune_demo/output
+finetune_demo/data
+finetune_demo/formatted_data
+ToolAlpaca/
+AdvertiseGen/
+*.gz
+*.idea
\ No newline at end of file
diff --git a/.ipynb_checkpoints/Untitled-checkpoint.ipynb b/.ipynb_checkpoints/Untitled-checkpoint.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..363fcab7ed6e9634e198cf5555ceb88932c9a245
--- /dev/null
+++ b/.ipynb_checkpoints/Untitled-checkpoint.ipynb
@@ -0,0 +1,6 @@
+{
+ "cells": [],
+ "metadata": {},
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/DEPLOYMENT.md b/DEPLOYMENT.md
new file mode 100644
index 0000000000000000000000000000000000000000..af008abe45e5e808ad99aa5963d9d640e3d77adb
--- /dev/null
+++ b/DEPLOYMENT.md
@@ -0,0 +1,39 @@
+## 低成本部署
+
+### 模型量化
+
+默认情况下，模型以 FP16 精度加载，运行上述代码需要大概 13GB 显存。如果你的 GPU 显存有限，可以尝试以量化方式加载模型，使用方法如下：
+
+```python
+model = AutoModel.from_pretrained("THUDM/chatglm3-6b",trust_remote_code=True).quantize(4).cuda()
+```
+
+模型量化会带来一定的性能损失，经过测试，ChatGLM3-6B 在 4-bit 量化下仍然能够进行自然流畅的生成。
+
+### CPU 部署
+
+如果你没有 GPU 硬件的话，也可以在 CPU 上进行推理，但是推理速度会更慢。使用方法如下（需要大概 32GB 内存）
+```python
+model = AutoModel.from_pretrained("THUDM/chatglm3-6b", trust_remote_code=True).float()
+```
+
+### Mac 部署
+
+对于搭载了 Apple Silicon 或者 AMD GPU 的 Mac，可以使用 MPS 后端来在 GPU 上运行 ChatGLM3-6B。需要参考 Apple 的 [官方说明](https://developer.apple.com/metal/pytorch) 安装 PyTorch-Nightly（正确的版本号应该是2.x.x.dev2023xxxx，而不是 2.x.x）。
+
+目前在 MacOS 上只支持[从本地加载模型](README.md#从本地加载模型)。将代码中的模型加载改为从本地加载，并使用 mps 后端：
+```python
+model = AutoModel.from_pretrained("your local path", trust_remote_code=True).to('mps')
+```
+
+加载半精度的 ChatGLM3-6B 模型需要大概 13GB 内存。内存较小的机器（比如 16GB 内存的 MacBook Pro），在空余内存不足的情况下会使用硬盘上的虚拟内存，导致推理速度严重变慢。
+
+### 多卡部署
+如果你有多张 GPU，但是每张 GPU 的显存大小都不足以容纳完整的模型，那么可以将模型切分在多张GPU上。首先安装 accelerate: `pip install accelerate`，然后通过如下方法加载模型：
+
+```python
+from utils import load_model_on_gpus
+
+model = load_model_on_gpus("THUDM/chatglm3-6b", num_gpus=2)
+```
+即可将模型部署到两张 GPU 上进行推理。你可以将 `num_gpus` 改为你希望使用的 GPU 数。默认是均匀切分的，你也可以传入 `device_map` 参数来自己指定。 
\ No newline at end of file
diff --git a/DEPLOYMENT_en.md b/DEPLOYMENT_en.md
new file mode 100644
index 0000000000000000000000000000000000000000..18cce66db358ce617d0f3c33aaf9ab5666139284
--- /dev/null
+++ b/DEPLOYMENT_en.md
@@ -0,0 +1,43 @@
+## Low-Cost Deployment
+
+### Model Quantization
+
+By default, the model is loaded with FP16 precision, running the above code requires about 13GB of VRAM. If your GPU's VRAM is limited, you can try loading the model quantitatively, as follows:
+
+```python
+model = AutoModel.from_pretrained("THUDM/chatglm3-6b",trust_remote_code=True).quantize(4).cuda()
+```
+
+Model quantization will bring some performance loss. Through testing, ChatGLM3-6B can still perform natural and smooth generation under 4-bit quantization.
+
+### CPU Deployment
+
+If you don't have GPU hardware, you can also run inference on the CPU, but the inference speed will be slower. The usage is as follows (requires about 32GB of memory):
+
+```python
+model = AutoModel.from_pretrained("THUDM/chatglm3-6b", trust_remote_code=True).float()
+```
+
+### Mac Deployment
+
+For Macs equipped with Apple Silicon or AMD GPUs, the MPS backend can be used to run ChatGLM3-6B on the GPU. Refer to Apple's [official instructions](https://developer.apple.com/metal/pytorch) to install PyTorch-Nightly (the correct version number should be 2.x.x.dev2023xxxx, not 2.x.x).
+
+Currently, only [loading the model locally](README_en.md#load-model-locally) is supported on MacOS. Change the model loading in the code to load locally and use the MPS backend:
+
+```python
+model = AutoModel.from_pretrained("your local path", trust_remote_code=True).to('mps')
+```
+
+Loading the half-precision ChatGLM3-6B model requires about 13GB of memory. Machines with smaller memory (such as a 16GB memory MacBook Pro) will use virtual memory on the hard disk when there is insufficient free memory, resulting in a significant slowdown in inference speed.
+
+### Multi-GPU Deployment
+
+If you have multiple GPUs, but each GPU's VRAM size is not enough to accommodate the complete model, then the model can be split across multiple GPUs. First, install accelerate: `pip install accelerate`, and then load the model through the following methods:
+
+```python
+from utils import load_model_on_gpus
+
+model = load_model_on_gpus("THUDM/chatglm3-6b", num_gpus=2)
+```
+
+This allows the model to be deployed on two GPUs for inference. You can change `num_gpus` to the number of GPUs you want to use. It is evenly split by default, but you can also pass the `device_map` parameter to specify it yourself.
\ No newline at end of file
diff --git a/MODEL_LICENSE b/MODEL_LICENSE
new file mode 100644
index 0000000000000000000000000000000000000000..70ddc809f396cd1d05827a86d159fb72553ff5a9
--- /dev/null
+++ b/MODEL_LICENSE
@@ -0,0 +1,65 @@
+The ChatGLM3-6B License
+
+1. 定义
+
+“许可方”是指分发其软件的 ChatGLM3-6B 模型团队。
+
+“软件”是指根据本许可提供的 ChatGLM3-6B 模型参数。
+
+2. 许可授予
+
+根据本许可的条款和条件，许可方特此授予您非排他性、全球性、不可转让、不可再许可、可撤销、免版税的版权许可。
+
+上述版权声明和本许可声明应包含在本软件的所有副本或重要部分中。
+
+3.限制
+
+您不得出于任何军事或非法目的使用、复制、修改、合并、发布、分发、复制或创建本软件的全部或部分衍生作品。
+
+您不得利用本软件从事任何危害国家安全和国家统一、危害社会公共利益、侵犯人身权益的行为。
+
+4.免责声明
+
+本软件“按原样”提供，不提供任何明示或暗示的保证，包括但不限于对适销性、特定用途的适用性和非侵权性的保证。 在任何情况下，作者或版权持有人均不对任何索赔、损害或其他责任负责，无论是在合同诉讼、侵权行为还是其他方面，由软件或软件的使用或其他交易引起、由软件引起或与之相关 软件。
+
+5. 责任限制
+
+除适用法律禁止的范围外，在任何情况下且根据任何法律理论，无论是基于侵权行为、疏忽、合同、责任或其他原因，任何许可方均不对您承担任何直接、间接、特殊、偶然、示范性、 或间接损害，或任何其他商业损失，即使许可人已被告知此类损害的可能性。
+
+6.争议解决
+
+本许可受中华人民共和国法律管辖并按其解释。 因本许可引起的或与本许可有关的任何争议应提交北京市海淀区人民法院。
+
+请注意，许可证可能会更新到更全面的版本。 有关许可和版权的任何问题，请通过 license@zhipuai.cn 与我们联系。
+
+1. Definitions
+
+“Licensor” means the ChatGLM3-6B Model Team that distributes its Software.
+
+“Software” means the ChatGLM3-6B model parameters made available under this license.
+
+2. License Grant
+
+Subject to the terms and conditions of this License, the Licensor hereby grants to you a non-exclusive, worldwide, non-transferable, non-sublicensable, revocable, royalty-free copyright license to use the Software.
+
+The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
+
+3. Restriction
+
+You will not use, copy, modify, merge, publish, distribute, reproduce, or create derivative works of the Software, in whole or in part, for any military, or illegal purposes.
+
+You will not use the Software for any act that may undermine China's national security and national unity, harm the public interest of society, or infringe upon the rights and interests of human beings.
+
+4. Disclaimer
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+5. Limitation of Liability
+
+EXCEPT TO THE EXTENT PROHIBITED BY APPLICABLE LAW, IN NO EVENT AND UNDER NO LEGAL THEORY, WHETHER BASED IN TORT, NEGLIGENCE, CONTRACT, LIABILITY, OR OTHERWISE WILL ANY LICENSOR BE LIABLE TO YOU FOR ANY DIRECT, INDIRECT, SPECIAL, INCIDENTAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES, OR ANY OTHER COMMERCIAL LOSSES, EVEN IF THE LICENSOR HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
+
+6. Dispute Resolution
+
+This license shall be governed and construed in accordance with the laws of People’s Republic of China. Any dispute arising from or in connection with this License shall be submitted to Haidian District People's Court in Beijing.
+
+Note that the license is subject to update to a more comprehensive version.  For any questions related to the license and copyright, please contact us at license@zhipuai.cn.
diff --git a/PROMPT.md b/PROMPT.md
new file mode 100644
index 0000000000000000000000000000000000000000..dd73d327261290dfd615ba82d54cbe0106f60c95
--- /dev/null
+++ b/PROMPT.md
@@ -0,0 +1,203 @@
+## ChatGLM3 对话格式
+为了避免用户输入的注入攻击，以及统一 Code Interpreter，Tool & Agent 等任务的输入，ChatGLM3 采用了全新的对话格式。
+
+### 规定
+#### 整体结构
+ChatGLM3 对话的格式由若干对话组成，其中每个对话包含对话头和内容，一个典型的多轮对话结构如下
+```text
+<|system|>
+You are ChatGLM3, a large language model trained by Zhipu.AI. Follow the user's instructions carefully. Respond using markdown.
+<|user|>
+Hello
+<|assistant|>
+Hello, I'm ChatGLM3. What can I assist you today?
+```
+**实际中每轮对话内容并不一定以换行符结尾，这里只是为了美观，下同**
+
+#### 对话头
+对话头占完整的一行，格式为
+```text
+<|role|>{metadata}
+```
+其中 `<|role|>` 部分使用 special token 表示，无法从文本形式被 tokenizer 编码以防止注入。metadata 部分采用纯文本表示，为可选内容。
+* `<|system|>`：系统信息，设计上可穿插于对话中，**但目前规定仅可以出现在开头**
+* `<|user|>`：用户
+  - 不会连续出现多个来自 `<|user|>` 的信息
+* `<|assistant|>`：AI 助手
+  - 在出现之前必须有一个来自 `<|user|>` 的信息
+* `<|observation|>`：外部的返回结果
+  - 必须在 `<|assistant|>` 的信息之后
+
+### 样例场景
+
+为提升可读性，下列样例场景中表示角色的 special token 前均额外添加了一个换行符。实际使用及 tokenizer 实现中均无需额外添加这一换行。
+
+#### 多轮对话
+* 有且仅有 `<|user|>`、`<|assistant|>`、`<|system|>` 三种 role
+```text
+<|system|>
+You are ChatGLM3, a large language model trained by Zhipu.AI. Follow the user's instructions carefully. Respond using markdown.
+<|user|>
+Hello
+<|assistant|>
+Hello, I'm ChatGLM3. What can I assist you today?
+```
+
+#### 工具调用
+````
+<|system|>
+Answer the following questions as best as you can. You have access to the following tools:
+[
+    {
+        "name": "get_current_weather",
+        "description": "Get the current weather in a given location",
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "location": {
+                    "type": "string",
+                    "description": "The city and state, e.g. San Francisco, CA",
+                },
+                "unit": {"type": "string"},
+            },
+            "required": ["location"],
+        },
+    }
+]
+<|user|>
+今天北京的天气怎么样？
+<|assistant|>
+好的，让我们来查看今天的天气
+<|assistant|>get_current_weather
+```python
+tool_call(location="beijing", unit="celsius")
+```
+<|observation|>
+{"temperature": 22}
+<|assistant|>
+根据查询结果，今天北京的气温为 22 摄氏度。
+````
+
+#### 代码执行
+* 有 `<|user|>`、`<|assistant|>`、`<|system|>`、`<|observation|>` 四种 role。其中 `<|assistant|>` 的 metadata 只有 interpreter。
+
+`````text
+<|system|>
+你是一位智能AI助手，你叫ChatGLM3，你连接着一台电脑，但请注意不能联网。在使用Python解决任务时，你可以运行代码并得到结果，如果运行结果有错误，你需要尽可能对代码进行改进。你可以处理用户上传到电脑上的文件，文件默认存储路径是/mnt/data/。
+<|user|>
+#File: /mnt/data/metadata.jsonl
+#Size: 35380
+#File uploaded
+文件中是否存在缺失值或异常值？
+<|assistant|>
+首先，我将读取您提供的文件，并查看其内容以确定是否存在缺失值或异常值。
+
+我们从加载文件开始：
+<|assistant|>interpreter
+```python
+import json
+
+# Load the uploaded file
+data = []
+with open("/mnt/data/metadata.jsonl", 'r') as file:
+    for line in file:
+        data.append(json.loads(line))
+
+# Display the first few records to get a sense of the data
+data[:5]
+```
+<|observation|>
+```result
+[{'file_name': 'aceinthehole.png',
+  'name': 'Ace in the Hole',
+  'type': 'survivor',
+  'description': 'Lady Luck always seems to be throwing something good your way.'},
+ {'file_name': 'adrenaline.png',
+  'name': 'Adrenaline',
+  'type': 'survivor',
+  'description': 'You are fuelled by unexpected energy when on the verge of escape.'},
+ {'file_name': 'aftercare.png',
+  'name': 'Aftercare',
+  'type': 'survivor',
+  'description': "Unlocks potential in one's aura reading ability."},
+ {'file_name': 'agitation.png',
+  'name': 'Agitation',
+  'type': 'killer',
+  'description': 'You get excited in anticipation of hooking your prey.'},
+ {'file_name': 'alert.png',
+  'name': 'Alert',
+  'type': 'survivor',
+  'description': 'Your acute senses are on high alert.'}]
+```
+<|assistant|>
+该文件看起来包含有关某些条目的元数据，每个条目有以下字段：
+
+
+- `file_name`: 文件名称
+- `name`: 名称
+- `type`: 类型（例如 "survivor" 或 "killer"）
+- `description`: 描述
+
+接下来，我将检查这些字段是否存在缺失值。
+<|assistant|>interpreter
+```python
+# Check for missing values in each column
+missing_values = {key: sum(1 for item in data if key not in item) for key in data[0].keys()}
+
+missing_values
+```
+<|observation|>
+```result
+{'file_name': 0, 'name': 0, 'type': 0, 'description': 0}
+```
+<|assistant|>
+每个字段都没有缺失值。这意味着每个条目都完整。
+
+接下来，我们可以检查是否存在任何潜在的异常值。这通常需要更深入的分析和对数据的了解，但我可以从一些基本的统计和分布开始进行检查。为了这样做，我们可以查看“type”字段的值分布，看看是否有任何意外的类别。
+<|assistant|>interpreter
+```python
+# Check the distribution of values in the 'type' column
+type_distribution = {key: sum(1 for item in data if item['type'] == key) for key in set(item['type'] for item in data)}
+
+type_distribution
+```
+<|observation|>
+```result
+{'survivor': 116, 'killer': 103}
+```
+<|assistant|>
+“type”字段有两个值：“survivor”和“killer”。其中，“survivor”有116个条目，而“killer”有103个条目。
+
+从这个初步的检查中，我们没有发现任何明显的异常值。
+
+如果您有任何特定的标准或其他字段需要检查，我可以为您提供更详细的分析。您还希望我检查什么呢？
+<|user|>
+帮我画一个爱心
+<|assistant|>interpreter
+```python
+import numpy as np
+import matplotlib.pyplot as plt
+
+# 爱心形状的参数方程
+def heart(t):
+    x = 16 * np.sin(t) ** 3
+    y = 13 * np.cos(t) - 5 * np.cos(2 * t) - 2 * np.cos(3 * t) - np.cos(4 * t)
+    return x, y
+
+t = np.linspace(0, 2 * np.pi, 1000)
+x, y = heart(t)
+
+plt.figure(figsize=(6, 6))
+plt.plot(x, y, color='red')
+plt.axis('equal')
+plt.axis('off')
+plt.show()
+```
+<|observation|>
+```result
+[Image]
+```
+<|assistant|>
+这是一个爱心形状。我使用了参数方程来描述这个形状，并使用matplotlib进行了绘制。如果您有任何其他需求或问题，请随时告诉我。
+<|user|>  # End
+````
\ No newline at end of file
diff --git a/PROMPT_en.md b/PROMPT_en.md
new file mode 100644
index 0000000000000000000000000000000000000000..bcfe2d516b3784dbb2e0f59d44bd506ec732802c
--- /dev/null
+++ b/PROMPT_en.md
@@ -0,0 +1,201 @@
+## ChatGLM3 Chat Format
+To avoid injection attacks from user input, and to unify the input of Code Interpreter, Tool & Agent and other tasks, ChatGLM3 adopts a brand-new dialogue format.
+
+### Regulations
+#### Overall Structure
+The format of the ChatGLM3 dialogue consists of several conversations, each of which contains a dialogue header and content. A typical multi-turn dialogue structure is as follows:
+```text
+<|system|>
+You are ChatGLM3, a large language model trained by Zhipu.AI. Follow the user's instructions carefully. Respond using markdown.
+<|user|>
+Hello
+<|assistant|>
+Hello, I'm ChatGLM3. What can I assist you today?
+```
+
+#### Chat Header
+The chat header occupies a complete line, formatted as:
+```text
+<|role|>{metadata}
+```
+Where `<|role|>` part is represented in a special token,  which can’t be encoded by the tokenizer from the text form to prevent injection attacks. The `metadata` part is represented in plain texts and is optional content.
+* `<|system|>`: System information, which can be interspersed in the dialogue in design, **but currently only appears at the beginning**
+* `<|user|>`: User
+  - Multiple messages from `<|user|>` will not appear continuously
+* `<|assistant|>`: AI assistant
+  - There must be a message from `<|user|>` before it appears
+* `<|observation|>`: External return result
+  - Must be after the message from `<|assistant|>`
+
+### Example Scenarios
+
+For better readablity, an extra `\n` is added before each role special token. This extra `\n` should not be added in actual use and tokenizer implementation.
+
+#### Multi-turn Dialogue
+* There are only three roles: `<|user|>`, `<|assistant|>`, and `<|system|>`.
+```text
+<|system|>
+You are ChatGLM3, a large language model trained by Zhipu.AI. Follow the user's instructions carefully. Respond using markdown.
+<|user|>
+Hello
+<|assistant|>
+Hello, I'm ChatGLM3. What can I assist you today?
+```
+
+#### Tool Calling
+````
+<|system|>
+Answer the following questions as best as you can. You have access to the following tools:
+[
+    {
+        "name": "get_current_weather",
+        "description": "Get the current weather in a given location",
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "location": {
+                    "type": "string",
+                    "description": "The city and state, e.g. San Francisco, CA",
+                },
+                "unit": {"type": "string"},
+            },
+            "required": ["location"],
+        },
+    }
+]
+<|user|>
+What's the weather in Beijing today?
+<|assistant|>
+Okay, let's look up the weather in Bejing today.
+<|assistant|>get_current_weather
+```python
+tool_call(location="beijing", unit="celsius")
+```
+<|observation|>
+{"temperature": 22}
+<|assistant|>
+According to the query results, the temperature in Beijing today is 22 degrees Celsius.
+````
+
+#### Code Execution
+* Four available roles: `<|user|>`, `<|assistant|>`, `<|system|>`, and `<|observation|>`, where `metadata` of `<|assistant|>` can only be interpreter.
+
+`````text
+<|system|>
+You are an intelligent AI assistant, named ChatGLM3. You are connected to a computer, but please note that you cannot connect to the internet. When using Python to solve tasks, you can run the code and get the results. If there are errors in the running results, you need to improve the code as much as possible. You can process files uploaded by users to your computer, and the default storage path for files is /mnt/data/.
+<|user|>
+#File: /mnt/data/metadata.jsonl
+#Size: 35380
+#File uploaded
+Are there missing or abnormal values in the file?
+<|assistant|>
+Firstly, I will read the file you provided and review its contents to determine if there are missing or abnormal values.
+
+We start by loading the file:
+<|assistant|>interpreter
+```python
+import json
+
+# Load the uploaded file
+data = []
+with open("/mnt/data/metadata.jsonl", 'r') as file:
+    for line in file:
+        data.append(json.loads(line))
+
+# Display the first few records to get a sense of the data
+data[:5]
+```
+<|observation|>
+```result
+[{'file_name': 'aceinthehole.png',
+  'name': 'Ace in the Hole',
+  'type': 'survivor',
+  'description': 'Lady Luck always seems to be throwing something good your way.'},
+ {'file_name': 'adrenaline.png',
+  'name': 'Adrenaline',
+  'type': 'survivor',
+  'description': 'You are fuelled by unexpected energy when on the verge of escape.'},
+ {'file_name': 'aftercare.png',
+  'name': 'Aftercare',
+  'type': 'survivor',
+  'description': "Unlocks potential in one's aura reading ability."},
+ {'file_name': 'agitation.png',
+  'name': 'Agitation',
+  'type': 'killer',
+  'description': 'You get excited in anticipation of hooking your prey.'},
+ {'file_name': 'alert.png',
+  'name': 'Alert',
+  'type': 'survivor',
+  'description': 'Your acute senses are on high alert.'}]
+```
+<|assistant|>
+This file appears to contain metadata about certain items, each with the following fields:
+
+- `file_name`
+- `name`
+- `type` (e.g. "survivor" or "killer")
+- `description`
+
+Next, I will check for missing values in these fields.
+<|assistant|>interpreter
+```python
+# Check for missing values in each column
+missing_values = {key: sum(1 for item in data if key not in item) for key in data[0].keys()}
+
+missing_values
+```
+<|observation|>
+```result
+{'file_name': 0, 'name': 0, 'type': 0, 'description': 0}
+```
+<|assistant|>
+Each field has no missing values. This means that each entry is complete.
+
+Next, we can check for any potential outliers. This usually requires deeper analysis and understanding of the data, but I can start by examining some basic statistics and distributions. To do this, we can check the value distribution of the "type" field to see if there are any unexpected categories.
+<|assistant|>interpreter
+```python
+# Check the distribution of values in the 'type' column
+type_distribution = {key: sum(1 for item in data if item['type'] == key) for key in set(item['type'] for item in data)}
+
+type_distribution
+```
+<|observation|>
+```result
+{'survivor': 116, 'killer': 103}
+```
+<|assistant|>
+The 'type' field has two values: 'survivor' and 'killer'. Among them, 'survivor' has 116 entries, while 'killer' has 103 entries.
+
+From this preliminary examination, we did not find any obvious abnormal values.
+
+If you have any specific standards or other fields that need to be checked, I can provide you with a more detailed analysis. What else do you want me to check?
+<|user|>
+Help me draw a heart
+<|assistant|>interpreter
+```python
+import numpy as np
+import matplotlib.pyplot as plt
+
+# The parametric equation for the shape of a heart
+def heart(t):
+    x = 16 * np.sin(t) ** 3
+    y = 13 * np.cos(t) - 5 * np.cos(2 * t) - 2 * np.cos(3 * t) - np.cos(4 * t)
+    return x, y
+
+t = np.linspace(0, 2 * np.pi, 1000)
+x, y = heart(t)
+
+plt.figure(figsize=(6, 6))
+plt.plot(x, y, color='red')
+plt.axis('equal')
+plt.axis('off')
+plt.show()
+```
+<|observation|>
+```result
+[Image]
+```
+<|assistant|>
+This is a shape of heart. I used parametric equations to describe this shape and drew it using Matplotlib. If you have any other needs or questions, please feel free to let me know.
+<|user|>  # End
+````
diff --git a/README_en.md b/README_en.md
new file mode 100644
index 0000000000000000000000000000000000000000..48cd260e8d528e7bd8f040a8b314e23f438d000b
--- /dev/null
+++ b/README_en.md
@@ -0,0 +1,250 @@
+# ChatGLM3
+
+<p align="center">
+🤗 <a href="https://huggingface.co/THUDM/chatglm3-6b" target="_blank">HF Repo</a> • 🤖 <a href="https://modelscope.cn/models/ZhipuAI/chatglm3-6b" target="_blank">ModelScope</a> • 📔 <a href="https://lslfd0slxc.feishu.cn/wiki/WvQbwIJ9tiPAxGk8ywDck6yfnof" target="_blank">Document</a> • 🐦 <a href="https://twitter.com/thukeg" target="_blank">Twitter</a> • 📃 <a href="https://arxiv.org/abs/2103.10360" target="_blank">[GLM@ACL 22]</a> <a href="https://github.com/THUDM/GLM" target="_blank">[GitHub]</a> • 📃 <a href="https://arxiv.org/abs/2210.02414" target="_blank">[GLM-130B@ICLR 23]</a> <a href="https://github.com/THUDM/GLM-130B" target="_blank">[GitHub]</a> <br>
+</p>
+<p align="center">
+    👋 Join our <a href="https://join.slack.com/t/chatglm/shared_invite/zt-25ti5uohv-A_hs~am_D3Q8XPZMpj7wwQ" target="_blank">Slack</a> and <a href="resources/WECHAT.md" target="_blank">WeChat</a>
+</p>
+<p align="center">
+📍Experience the larger-scale ChatGLM model at <a href="https://www.chatglm.cn">chatglm.cn</a>
+</p>
+
+## Introduction
+
+ChatGLM3 is a new generation of pre-trained dialogue models jointly released by Zhipu AI and Tsinghua KEG. ChatGLM3-6B is the open-source model in the ChatGLM3 series, maintaining many excellent features of the first two generations such as smooth dialogue and low deployment threshold, while introducing the following features:
+
+1. **Stronger Base Model:** The base model of ChatGLM3-6B, ChatGLM3-6B-Base, adopts a more diverse training dataset, more sufficient training steps, and a more reasonable training strategy. Evaluations on datasets from various perspectives such as semantics, mathematics, reasoning, code, and knowledge show that **ChatGLM3-6B-Base has the strongest performance among base models below 10B**.
+
+2. **More Complete Function Support:** ChatGLM3-6B adopts a newly designed [Prompt format](PROMPT_en.md), supporting multi-turn dialogues as usual. It also natively supports [tool invocation](tool_using/README_en.md) (Function Call), code execution (Code Interpreter), and Agent tasks in complex scenarios.
+
+3. **More Comprehensive Open-source Series:** In addition to the dialogue model [ChatGLM3-6B](https://huggingface.co/THUDM/chatglm3-6b), the basic model [ChatGLM3-6B-Base](https://huggingface.co/THUDM/chatglm3-6b-base), and the long-text dialogue model [ChatGLM3-6B-32K](https://huggingface.co/THUDM/chatglm3-6b-32k) have also been open-sourced. All these weights are **fully open** for academic research, and **free commercial use is also allowed** after registration via a [questionnaire](https://open.bigmodel.cn/mla/form).
+
+-----
+
+The ChatGLM3 open-source model aims to promote the development of large-model technology together with the open-source community. Developers and everyone are earnestly requested to comply with the [open-source protocol](MODEL_LICENSE), and not to use the open-source models, codes, and derivatives for any purposes that might harm the nation and society, and for any services that have not been evaluated and filed for safety. Currently, no applications, including web, Android, Apple iOS, and Windows App, have been developed based on the **ChatGLM3 open-source model** by our project team.
+
+Although every effort has been made to ensure the compliance and accuracy of the data at various stages of model training, due to the smaller scale of the ChatGLM3-6B model and the influence of probabilistic randomness factors, the accuracy of output content cannot be guaranteed. The model output is also easily misled by user input. **This project does not assume risks and liabilities caused by data security, public opinion risks, or any misleading, abuse, dissemination, and improper use of open-source models and codes.**
+
+## Model List
+
+| Model | Seq Length |                                                              Download                                                               
+| :---: |:---------------------------:|:-----------------------------------------------------------------------------------------------------------------------------------:
+| ChatGLM3-6B | 8k |      [HuggingFace](https://huggingface.co/THUDM/chatglm3-6b) \| [ModelScope](https://modelscope.cn/models/ZhipuAI/chatglm3-6b)      
+| ChatGLM3-6B-Base | 8k | [HuggingFace](https://huggingface.co/THUDM/chatglm3-6b-base) \| [ModelScope](https://modelscope.cn/models/ZhipuAI/chatglm3-6b-base) 
+| ChatGLM3-6B-32K | 32k |                                   [HuggingFace](https://huggingface.co/THUDM/chatglm3-6b-32k) \| [ModelScope](https://modelscope.cn/models/ZhipuAI/chatglm3-6b-32k)                                    
+
+## Projects
+Open source projects that accelerate ChatGLM3:
+* [chatglm.cpp](https://github.com/li-plus/chatglm.cpp): Real-time inference on your laptop accelerated by quantization, similar to llama.cpp.
+* [ChatGLM3-TPU](https://github.com/sophgo/ChatGLM3-TPU): Using the TPU accelerated inference solution, it runs about 7.5 token/s in real time on the end-side chip BM1684X (16T@FP16, 16G DDR).
+
+## Evaluation Results
+
+### Typical Tasks
+
+We selected 8 typical Chinese-English datasets and conducted performance tests on the ChatGLM3-6B (base) version.
+
+| Model            | GSM8K | MATH | BBH  | MMLU | C-Eval | CMMLU | MBPP | AGIEval |
+|------------------|:-----:|:----:|:----:|:----:|:------:|:-----:|:----:|:-------:|
+| ChatGLM2-6B-Base | 32.4  | 6.5  | 33.7 | 47.9 |  51.7  | 50.0  |  -   |    -    |
+| Best Baseline    | 52.1  | 13.1 | 45.0 | 60.1 |  63.5  | 62.2  | 47.5 |  45.8   |
+| ChatGLM3-6B-Base | 72.3  | 25.7 | 66.1 | 61.4 |  69.0  | 67.5  | 52.4 |  53.7   |
+> "Best Baseline" refers to the pre-trained models that perform best on the corresponding datasets with model parameters below 10B, excluding models that are trained specifically for a single task and do not maintain general capabilities.
+
+> In the tests of ChatGLM3-6B-Base, BBH used a 3-shot test, GSM8K and MATH that require inference used a 0-shot CoT test, MBPP used a 0-shot generation followed by running test cases to calculate Pass@1, and other multiple-choice type datasets all used a 0-shot test.
+
+We have conducted manual evaluation tests on ChatGLM3-6B-32K in multiple long-text application scenarios. Compared with the second-generation model, its effect has improved by more than 50% on average. In applications such as paper reading, document summarization, and financial report analysis, this improvement is particularly significant. In addition, we also tested the model on the LongBench evaluation set, and the specific results are shown in the table below.
+
+| Model                |  Average |  Summary | Single-Doc QA |  Multi-Doc QA | Code | Few-shot | Synthetic | 
+|----------------------|:-----:|:----:|:----:|:----:|:------:|:-----:|:-----:|
+| ChatGLM2-6B-32K   |  41.5 | 24.8 | 37.6 | 34.7 |  52.8  |  51.3 | 47.7 | 
+| ChatGLM3-6B-32K   |  50.2 | 26.6 | 45.8 | 46.1 |  56.2  |  61.2 | 65 |
+
+
+## How to Use
+
+### Environment Installation
+First, you need to download this repository:
+```shell
+git clone https://github.com/THUDM/ChatGLM3
+cd ChatGLM3
+```
+
+Then use pip to install the dependencies:
+```
+pip install -r requirements.txt
+```
++ The `transformers` library version should be `4.30.2` and above, and `torch` library should be 2.0 and above to obtain the best inference performance.
++ In order to ensure that the version of `torch` is correct, please strictly follow the instructions of [official documentation](https://pytorch.org/get-started/locally/) for installation.
++ The `gradio` library version should be the `3.x` version.
+
+
+### Integrated Demo
+
+We provide an integrated demo that incorporates the following three functionalities. Please refer to [Integrated Demo](composite_demo/README_en.md) for how to run it.
+
+- Chat: Dialogue mode, where you can interact with the model.
+- Tool: Tool mode, where in addition to dialogue, the model can also perform other operations using tools.
+    ![tool](resources/tool_en.png)
+- Code Interpreter: Code interpreter mode, where the model can execute code in a Jupyter environment and obtain results to complete complex tasks.
+    ![code](resources/code_en.gif)
+
+### Usage 
+
+The ChatGLM model can be called to start a conversation using the following code:
+
+```python
+>>> from transformers import AutoTokenizer, AutoModel
+>>> tokenizer = AutoTokenizer.from_pretrained("THUDM/chatglm3-6b", trust_remote_code=True)
+>>> model = AutoModel.from_pretrained("THUDM/chatglm3-6b", trust_remote_code=True, device='cuda')
+>>> model = model.eval()
+>>> response, history = model.chat(tokenizer, "Hello", history=[])
+>>> print(response)
+Hello 👋! I'm ChatGLM3-6B, the artificial intelligence assistant, nice to meet you. Feel free to ask me any questions.
+>>> response, history = model.chat(tokenizer, "What should I do if I can't sleep at night", history=history)
+>>> print(response)
+If you're having trouble sleeping at night, here are a few suggestions that might help:
+
+1. Create a relaxing sleep environment: Make sure your bedroom is cool, quiet, and dark. Consider using earplugs, a white noise machine, or a fan to help create an optimal environment.
+2. Establish a bedtime routine: Try to go to bed and wake up at the same time every day, even on weekends. A consistent routine can help regulate your body's internal clock.
+3. Avoid stimulating activities before bedtime: Avoid using electronic devices, watching TV, or engaging in stimulating activities like exercise or puzzle-solving, as these can interfere with your ability to fall asleep.
+4. Limit caffeine and alcohol: Avoid consuming caffeine and alcohol close to bedtime, as these can disrupt your sleep patterns.
+5. Practice relaxation techniques: Try meditation, deep breathing, or progressive muscle relaxation to help calm your mind and body before sleep.
+6. Consider taking a warm bath or shower: A warm bath or shower can help relax your muscles and promote sleep.
+7. Get some fresh air: Make sure to get some fresh air during the day, as lack of vitamin D can interfere with sleep quality.
+
+If you continue to have difficulty sleeping, consult with a healthcare professional for further guidance and support.
+```
+
+#### Load Model Locally
+The above code will automatically download the model implementation and parameters by `transformers`. The complete model implementation is available on [Hugging Face Hub](https://huggingface.co/THUDM/chatglm3-6b). If your network environment is poor, downloading model parameters might take a long time or even fail. In this case, you can first download the model to your local machine, and then load it from there.
+
+To download the model from Hugging Face Hub, you need to [install Git LFS](https://docs.github.com/en/repositories/working-with-files/managing-large-files/installing-git-large-file-storage) first, then run
+```Shell
+git clone https://huggingface.co/THUDM/chatglm3-6b
+```
+
+If the download from HuggingFace is slow, you can also download it from [ModelScope](https://modelscope.cn/models/ZhipuAI/chatglm3-6b).
+
+# Model Fine-tuning
+Please refer to the dialog model fine-tuning [ChatGLM3-6B fine-tuning example](finetune_chatmodel_demo/README.md), or the base model fine-tuning [ChatGLM3-6B-base fine-tuning example](finetune_basemodel_demo/README.md).
+Please note that different fine-tuning scripts correspond to different models. Please select the corresponding model according to your needs.
+
+### Web-based Dialogue Demo
+![web-demo](resources/web-demo.gif)
+You can launch a web-based demo using Gradio with the following command:
+```shell
+python web_demo.py
+```
+
+![web-demo](resources/web-demo2.png)
+
+You can launch a web-based demo using Streamlit with the following command:
+```shell
+streamlit run web_demo2.py
+```
+
+The web-based demo will run a Web Server and output an address. You can use it by opening the output address in a browser. Based on tests, the web-based demo using Streamlit runs more smoothly.
+
+### Command Line Dialogue Demo
+
+![cli-demo](resources/cli-demo.png)
+
+Run [cli_demo.py](basic_demo/cli_demo.py) in the repository:
+
+```shell
+python cli_demo.py
+```
+
+The program will interact in the command line, enter instructions in the command line and hit enter to generate a response. Enter `clear` to clear the dialogue history, enter `stop` to terminate the program.
+
+### API Deployment
+Thanks to [@xusenlinzy](https://github.com/xusenlinzy) for implementing the OpenAI format streaming API deployment, which can serve as the backend for any ChatGPT-based application, such as [ChatGPT-Next-Web](https://github.com/Yidadaa/ChatGPT-Next-Web). You can deploy it by running [openai_api.py](openai_api_demo/openai_api.py) in the repository:
+```shell
+cd openai_api_demo
+python openai_api.py
+```
+Also, we have written a sample code to test the performance of the API calls. This can be tested by running [openai_api_request.py](openai_api_demo/openai_api_request.py) in the repository
++ Test with Curl
+```shell
+curl -X POST "http://127.0.0.1:8000/v1/chat/completions" \
+-H "Content-Type: application/json" \\
+-d "{\"model\": \"chatglm3-6b\", \"messages\": [{\"role\": \"system\", \"content\": \"You are ChatGLM3, a large language model trained by Zhipu. Follow the user's instructions carefully. Respond using markdown.\"}, {\"role\": \"user\", \"content\": \"Hello, tell me a story, about 100 words\"}], \"stream\": false, \"max_title": \"\". false, \"max_tokens\": 100, \"temperature\": 0.8, \"top_p\": 0.8}"
+````
++ Testing with Python
+```shell
+cd openai_api_demo
+python openai_api_request.py
+```
+If the test is successful, the model should return a story.
+
+### Tool Invocation
+
+For methods of tool invocation, please refer to [Tool Invocation](tool_using/README_en.md).
+
+## Low-Cost Deployment
+
+### Model Quantization
+
+By default, the model is loaded with FP16 precision, running the above code requires about 13GB of VRAM. If your GPU's VRAM is limited, you can try loading the model quantitatively, as follows:
+
+```python
+model = AutoModel.from_pretrained("THUDM/chatglm3-6b",trust_remote_code=True).quantize(4).cuda()
+```
+
+Model quantization will bring some performance loss. Through testing, ChatGLM3-6B can still perform natural and smooth generation under 4-bit quantization.
+
+### CPU Deployment
+
+If you don't have GPU hardware, you can also run inference on the CPU, but the inference speed will be slower. The usage is as follows (requires about 32GB of memory):
+
+```python
+model = AutoModel.from_pretrained("THUDM/chatglm3-6b", trust_remote_code=True).float()
+```
+
+### Mac Deployment
+
+For Macs equipped with Apple Silicon or AMD GPUs, the MPS backend can be used to run ChatGLM3-6B on the GPU. Refer to Apple's [official instructions](https://developer.apple.com/metal/pytorch) to install PyTorch-Nightly (the correct version number should be 2.x.x.dev2023xxxx, not 2.x.x).
+
+Currently, only [loading the model locally](README_en.md#load-model-locally) is supported on MacOS. Change the model loading in the code to load locally and use the MPS backend:
+
+```python
+model = AutoModel.from_pretrained("your local path", trust_remote_code=True).to('mps')
+```
+
+Loading the half-precision ChatGLM3-6B model requires about 13GB of memory. Machines with smaller memory (such as a 16GB memory MacBook Pro) will use virtual memory on the hard disk when there is insufficient free memory, resulting in a significant slowdown in inference speed.
+
+### Multi-GPU Deployment
+
+If you have multiple GPUs, but each GPU's VRAM size is not enough to accommodate the complete model, then the model can be split across multiple GPUs. First, install accelerate: `pip install accelerate`, and then load the model through the following methods:
+
+```python
+from utils import load_model_on_gpus
+
+model = load_model_on_gpus("THUDM/chatglm3-6b", num_gpus=2)
+```
+
+This allows the model to be deployed on two GPUs for inference. You can change `num_gpus` to the number of GPUs you want to use. It is evenly split by default, but you can also pass the `device_map` parameter to specify it yourself.
+
+## Citation
+
+If you find our work helpful, please consider citing the following papers.
+
+```
+@article{zeng2022glm,
+  title={Glm-130b: An open bilingual pre-trained model},
+  author={Zeng, Aohan and Liu, Xiao and Du, Zhengxiao and Wang, Zihan and Lai, Hanyu and Ding, Ming and Yang, Zhuoyi and Xu, Yifan and Zheng, Wendi and Xia, Xiao and others},
+  journal={arXiv preprint arXiv:2210.02414},
+  year={2022}
+}
+```
+```
+@inproceedings{du2022glm,
+  title={GLM: General Language Model Pretraining with Autoregressive Blank Infilling},
+  author={Du, Zhengxiao and Qian, Yujie and Liu, Xiao and Ding, Ming and Qiu, Jiezhong and Yang, Zhilin and Tang, Jie},
+  booktitle={Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
+  pages={320--335},
+  year={2022}
+}
+```
diff --git a/Untitled.ipynb b/Untitled.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..09af111c06dab456c78653d6779f100b9513d6bd
--- /dev/null
+++ b/Untitled.ipynb
@@ -0,0 +1,192 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "d1389f08-db10-44c0-acc0-08170ac549d0",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Requirement already satisfied: huggingface_hub in /opt/conda/lib/python3.10/site-packages (0.19.4)\n",
+      "Requirement already satisfied: filelock in /opt/conda/lib/python3.10/site-packages (from huggingface_hub) (3.9.0)\n",
+      "Requirement already satisfied: fsspec>=2023.5.0 in /opt/conda/lib/python3.10/site-packages (from huggingface_hub) (2023.10.0)\n",
+      "Requirement already satisfied: requests in /opt/conda/lib/python3.10/site-packages (from huggingface_hub) (2.31.0)\n",
+      "Requirement already satisfied: tqdm>=4.42.1 in /opt/conda/lib/python3.10/site-packages (from huggingface_hub) (4.65.0)\n",
+      "Requirement already satisfied: pyyaml>=5.1 in /opt/conda/lib/python3.10/site-packages (from huggingface_hub) (6.0.1)\n",
+      "Requirement already satisfied: typing-extensions>=3.7.4.3 in /opt/conda/lib/python3.10/site-packages (from huggingface_hub) (4.8.0)\n",
+      "Requirement already satisfied: packaging>=20.9 in /opt/conda/lib/python3.10/site-packages (from huggingface_hub) (23.1)\n",
+      "Requirement already satisfied: charset-normalizer<4,>=2 in /opt/conda/lib/python3.10/site-packages (from requests->huggingface_hub) (2.0.4)\n",
+      "Requirement already satisfied: idna<4,>=2.5 in /opt/conda/lib/python3.10/site-packages (from requests->huggingface_hub) (3.4)\n",
+      "Requirement already satisfied: urllib3<3,>=1.21.1 in /opt/conda/lib/python3.10/site-packages (from requests->huggingface_hub) (1.26.18)\n",
+      "Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.10/site-packages (from requests->huggingface_hub) (2023.7.22)\n",
+      "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
+      "\u001b[0mNote: you may need to restart the kernel to use updated packages.\n"
+     ]
+    }
+   ],
+   "source": [
+    "pip install huggingface_hub"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "37ce339b-5f87-4268-a808-7ec1f9b04055",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "e74cef2b1046454abf03203299de8bd7",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "VBox(children=(HTML(value='<center> <img\\nsrc=https://huggingface.co/front/assets/huggingface_logo-noborder.sv…"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from huggingface_hub import login\n",
+    "login()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "b338999a-2f2e-45fd-8bfe-ff5c035e79f3",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "ff4e7de5abd24397a2476b50df747fc0",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "pytorch_model-00002-of-00007.bin:   0%|          | 0.00/1.97G [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4810a0b8ac7c4537ae3b825eaaacd03d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Upload 11 LFS files:   0%|          | 0/11 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "2c378790ca33452bac16db72d1a93b5d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "pytorch_model-00005-of-00007.bin:   0%|          | 0.00/1.97G [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "19aa1f962f30403d8487851207ebc11f",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "pytorch_model-00003-of-00007.bin:   0%|          | 0.00/1.93G [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "700a7f07898f476c84dee0481d5dd594",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "pytorch_model-00001-of-00007.bin:   0%|          | 0.00/1.83G [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "0dbd5185b81f4cb380005865a8036e8b",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "pytorch_model-00004-of-00007.bin:   0%|          | 0.00/1.82G [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from huggingface_hub import HfApi\n",
+    "api = HfApi()\n",
+    "\n",
+    "api.upload_folder(\n",
+    "    folder_path=\"\",\n",
+    "    repo_id=\"kevinwang676/ChatGLM3-demo\",\n",
+    "    repo_type=\"space\",\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "ff1ddc96-5925-4978-bb9c-d016b716ade5",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.13"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/basic_demo/cli_demo.py b/basic_demo/cli_demo.py
new file mode 100644
index 0000000000000000000000000000000000000000..76bda5acb63c8206c2ab7ec88538907b86193709
--- /dev/null
+++ b/basic_demo/cli_demo.py
@@ -0,0 +1,79 @@
+import os
+import platform
+from transformers import AutoTokenizer, AutoModel
+import torch
+
+MODEL_PATH = os.environ.get('MODEL_PATH', 'THUDM/chatglm3-6b')
+TOKENIZER_PATH = os.environ.get("TOKENIZER_PATH", MODEL_PATH)
+DEVICE = 'cuda' if torch.cuda.is_available() else 'cpu'
+
+# for Mac Computer like M1
+# You Need Use Pytorch compiled with Metal
+# DEVICE = 'mps'
+
+# for AMD gpu likes MI100 (Not Official Steady Support yet)
+# You Need Use Pytorch compiled with ROCm
+# DEVICE = 'cuda'
+
+# for Intel gpu likes A770 (Not Official Steady Support yet)
+# You Need Use Pytorch compiled with oneDNN and install intel-extension-for-pytorch
+# import intel_extension_for_pytorch as ipex
+# DEVICE = 'xpu'
+
+# for Moore Threads gpu like MTT S80 (Not Official Steady Support yet)
+# You Need Use Pytorch compiled with Musa
+# DEVICE = 'musa'
+
+
+
+tokenizer = AutoTokenizer.from_pretrained(TOKENIZER_PATH, trust_remote_code=True)
+if 'cuda' in DEVICE: # AMD, NVIDIA GPU can use Half Precision
+    model = AutoModel.from_pretrained(MODEL_PATH, trust_remote_code=True).to(DEVICE).eval()
+else: # CPU, Intel GPU and other GPU can use Float16 Precision Only
+    model = AutoModel.from_pretrained(MODEL_PATH, trust_remote_code=True).float().to(DEVICE).eval()
+
+os_name = platform.system()
+clear_command = 'cls' if os_name == 'Windows' else 'clear'
+stop_stream = False
+
+welcome_prompt = "欢迎使用 ChatGLM3-6B 模型，输入内容即可进行对话，clear 清空对话历史，stop 终止程序"
+
+
+def build_prompt(history):
+    prompt = welcome_prompt
+    for query, response in history:
+        prompt += f"\n\n用户：{query}"
+        prompt += f"\n\nChatGLM3-6B：{response}"
+    return prompt
+
+
+def main():
+    past_key_values, history = None, []
+    global stop_stream
+    print(welcome_prompt)
+    while True:
+        query = input("\n用户：")
+        if query.strip() == "stop":
+            break
+        if query.strip() == "clear":
+            past_key_values, history = None, []
+            os.system(clear_command)
+            print(welcome_prompt)
+            continue
+        print("\nChatGLM：", end="")
+        current_length = 0
+        for response, history, past_key_values in model.stream_chat(tokenizer, query, history=history, top_p=1,
+                                                                    temperature=0.01,
+                                                                    past_key_values=past_key_values,
+                                                                    return_past_key_values=True):
+            if stop_stream:
+                stop_stream = False
+                break
+            else:
+                print(response[current_length:], end="", flush=True)
+                current_length = len(response)
+        print("")
+
+
+if __name__ == "__main__":
+    main()
diff --git a/basic_demo/cli_demo_bad_word_ids.py b/basic_demo/cli_demo_bad_word_ids.py
new file mode 100644
index 0000000000000000000000000000000000000000..c799d2fd8d365cb31fcaf3bac0aeb8cb5842d9ec
--- /dev/null
+++ b/basic_demo/cli_demo_bad_word_ids.py
@@ -0,0 +1,83 @@
+"""
+This script demonstrates how to use the `bad_words_ids` argument to filter out.
+"""
+import os
+import platform
+from transformers import AutoTokenizer, AutoModel
+import torch
+
+MODEL_PATH = os.environ.get('MODEL_PATH', 'THUDM/chatglm3-6b')
+TOKENIZER_PATH = os.environ.get("TOKENIZER_PATH", MODEL_PATH)
+DEVICE = 'cuda' if torch.cuda.is_available() else 'cpu'
+
+tokenizer = AutoTokenizer.from_pretrained(TOKENIZER_PATH, trust_remote_code=True)
+if 'cuda' in DEVICE: # AMD, NVIDIA GPU can use Half Precision
+    model = AutoModel.from_pretrained(MODEL_PATH, trust_remote_code=True).to(DEVICE).eval()
+else: # CPU, Intel GPU and other GPU can use Float16 Precision Only
+    model = AutoModel.from_pretrained(MODEL_PATH, trust_remote_code=True).float().to(DEVICE).eval()
+
+os_name = platform.system()
+clear_command = 'cls' if os_name == 'Windows' else 'clear'
+stop_stream = False
+
+welcome_prompt = "欢迎使用 ChatGLM3-6B 模型，输入内容即可进行对话，clear 清空对话历史，stop 终止程序"
+
+# 定义不希望出现的词汇, 你可以自定义, 在这个例子中，如果模型回答包含 "你好" 或 "ChatGLM"，则会出现这个报错
+# probability tensor contains either `inf`, `nan` or element < 0
+
+bad_words = ["你好", "ChatGLM"]
+# 将这些词汇转换为token ID列表，每个短语是一个子列表
+bad_word_ids = [tokenizer.encode(bad_word, add_special_tokens=False) for bad_word in bad_words]
+
+
+def build_prompt(history):
+    prompt = welcome_prompt
+    for query, response in history:
+        prompt += f"\n\n用户：{query}"
+        prompt += f"\n\nChatGLM3-6B：{response}"
+    return prompt
+
+def main():
+    past_key_values, history = None, []
+    global stop_stream
+    print(welcome_prompt)
+    while True:
+        query = input("\n用户：")
+        if query.strip().lower() == "stop":
+            break
+        if query.strip().lower() == "clear":
+            past_key_values, history = None, []
+            os.system(clear_command)
+            print(welcome_prompt)
+            continue
+
+        # Attempt to generate a response
+        try:
+            print("\nChatGLM：", end="")
+            current_length = 0
+            response_generated = False
+            for response, history, past_key_values in model.stream_chat(
+                tokenizer, query, history=history, top_p=1,
+                temperature=0.01,
+                past_key_values=past_key_values,
+                return_past_key_values=True,
+                bad_words_ids=bad_word_ids  # assuming this is implemented correctly
+            ):
+                response_generated = True
+                # Check if the response contains any bad words
+                if any(bad_word in response for bad_word in bad_words):
+                    print("我的回答涉嫌了bad word")
+                    break  # Break the loop if a bad word is detected
+
+                # Otherwise, print the generated response
+                print(response[current_length:], end="", flush=True)
+                current_length = len(response)
+            if not response_generated:
+                print("没有生成任何回答。")
+        except RuntimeError as e:
+            print(f"生成文本时发生错误：{e}，这可能是涉及到设定的敏感词汇")
+
+        print("")
+
+if __name__ == "__main__":
+    main()
\ No newline at end of file
diff --git a/basic_demo/utils.py b/basic_demo/utils.py
new file mode 100644
index 0000000000000000000000000000000000000000..5b8275fd0ff2a888e04270c03dc9e5c212555c58
--- /dev/null
+++ b/basic_demo/utils.py
@@ -0,0 +1,58 @@
+import os
+from typing import Dict, Union, Optional
+from torch.nn import Module
+from transformers import AutoModel
+
+
+def auto_configure_device_map(num_gpus: int) -> Dict[str, int]:
+    # transformer.word_embeddings 占用1层
+    # transformer.final_layernorm 和 lm_head 占用1层
+    # transformer.layers 占用 28 层
+    # 总共30层分配到num_gpus张卡上
+    num_trans_layers = 28
+    per_gpu_layers = 30 / num_gpus
+
+    # bugfix: 在linux中调用torch.embedding传入的weight,input不在同一device上,导致RuntimeError
+    # windows下 model.device 会被设置成 transformer.word_embeddings.device
+    # linux下 model.device 会被设置成 lm_head.device
+    # 在调用chat或者stream_chat时,input_ids会被放到model.device上
+    # 如果transformer.word_embeddings.device和model.device不同,则会导致RuntimeError
+    # 因此这里将transformer.word_embeddings,transformer.final_layernorm,lm_head都放到第一张卡上
+    # 本文件来源于https://github.com/THUDM/ChatGLM-6B/blob/main/utils.py
+    # 仅此处做少许修改以支持ChatGLM3
+    device_map = {
+        'transformer.embedding.word_embeddings': 0,
+        'transformer.encoder.final_layernorm': 0,
+        'transformer.output_layer': 0,
+        'transformer.rotary_pos_emb': 0,
+        'lm_head': 0
+    }
+
+    used = 2
+    gpu_target = 0
+    for i in range(num_trans_layers):
+        if used >= per_gpu_layers:
+            gpu_target += 1
+            used = 0
+        assert gpu_target < num_gpus
+        device_map[f'transformer.encoder.layers.{i}'] = gpu_target
+        used += 1
+
+    return device_map
+
+
+def load_model_on_gpus(checkpoint_path: Union[str, os.PathLike], num_gpus: int = 2,
+                       device_map: Optional[Dict[str, int]] = None, **kwargs) -> Module:
+    if num_gpus < 2 and device_map is None:
+        model = AutoModel.from_pretrained(checkpoint_path, trust_remote_code=True, **kwargs).half().cuda()
+    else:
+        from accelerate import dispatch_model
+
+        model = AutoModel.from_pretrained(checkpoint_path, trust_remote_code=True, **kwargs).half()
+
+        if device_map is None:
+            device_map = auto_configure_device_map(num_gpus)
+
+        model = dispatch_model(model, device_map=device_map)
+
+    return model
\ No newline at end of file
diff --git a/basic_demo/web_demo.py b/basic_demo/web_demo.py
new file mode 100644
index 0000000000000000000000000000000000000000..a6e4e19ecfb54c5f05d5cf4c0b0263d31e3675e7
--- /dev/null
+++ b/basic_demo/web_demo.py
@@ -0,0 +1,117 @@
+import os
+from transformers import AutoModel, AutoTokenizer
+import gradio as gr
+import mdtex2html
+from utils import load_model_on_gpus
+import torch
+
+MODEL_PATH = os.environ.get('MODEL_PATH', 'THUDM/chatglm3-6b')
+TOKENIZER_PATH = os.environ.get("TOKENIZER_PATH", MODEL_PATH)
+DEVICE = 'cuda' if torch.cuda.is_available() else 'cpu'
+
+tokenizer = AutoTokenizer.from_pretrained(TOKENIZER_PATH, trust_remote_code=True)
+if 'cuda' in DEVICE: # AMD, NVIDIA GPU can use Half Precision
+    model = AutoModel.from_pretrained(MODEL_PATH, trust_remote_code=True).to(DEVICE).eval()
+else: # CPU, Intel GPU and other GPU can use Float16 Precision Only
+    model = AutoModel.from_pretrained(MODEL_PATH, trust_remote_code=True).float().to(DEVICE).eval()
+
+# 多显卡支持，使用下面两行代替上面一行，将num_gpus改为你实际的显卡数量
+# from utils import load_model_on_gpus
+# model = load_model_on_gpus("THUDM/chatglm3-6b", num_gpus=2)
+
+"""Override Chatbot.postprocess"""
+
+def postprocess(self, y):
+    if y is None:
+        return []
+    for i, (message, response) in enumerate(y):
+        y[i] = (
+            None if message is None else mdtex2html.convert((message)),
+            None if response is None else mdtex2html.convert(response),
+        )
+    return y
+
+
+gr.Chatbot.postprocess = postprocess
+
+
+def parse_text(text):
+    """copy from https://github.com/GaiZhenbiao/ChuanhuChatGPT/"""
+    lines = text.split("\n")
+    lines = [line for line in lines if line != ""]
+    count = 0
+    for i, line in enumerate(lines):
+        if "```" in line:
+            count += 1
+            items = line.split('`')
+            if count % 2 == 1:
+                lines[i] = f'<pre><code class="language-{items[-1]}">'
+            else:
+                lines[i] = f'<br></code></pre>'
+        else:
+            if i > 0:
+                if count % 2 == 1:
+                    line = line.replace("`", "\`")
+                    line = line.replace("<", "&lt;")
+                    line = line.replace(">", "&gt;")
+                    line = line.replace(" ", "&nbsp;")
+                    line = line.replace("*", "&ast;")
+                    line = line.replace("_", "&lowbar;")
+                    line = line.replace("-", "&#45;")
+                    line = line.replace(".", "&#46;")
+                    line = line.replace("!", "&#33;")
+                    line = line.replace("(", "&#40;")
+                    line = line.replace(")", "&#41;")
+                    line = line.replace("$", "&#36;")
+                lines[i] = "<br>" + line
+    text = "".join(lines)
+    return text
+
+
+def predict(input, chatbot, max_length, top_p, temperature, history, past_key_values):
+    chatbot.append((parse_text(input), ""))
+    for response, history, past_key_values in model.stream_chat(tokenizer, input, history,
+                                                                past_key_values=past_key_values,
+                                                                return_past_key_values=True,
+                                                                max_length=max_length, top_p=top_p,
+                                                                temperature=temperature):
+        chatbot[-1] = (parse_text(input), parse_text(response))
+
+        yield chatbot, history, past_key_values
+
+
+def reset_user_input():
+    return gr.update(value='')
+
+
+def reset_state():
+    return [], [], None
+
+
+with gr.Blocks() as demo:
+    gr.HTML("""<h1 align="center">ChatGLM3-6B</h1>""")
+
+    chatbot = gr.Chatbot()
+    with gr.Row():
+        with gr.Column(scale=4):
+            with gr.Column(scale=12):
+                user_input = gr.Textbox(show_label=False, placeholder="Input...", lines=10).style(
+                    container=False)
+            with gr.Column(min_width=32, scale=1):
+                submitBtn = gr.Button("Submit", variant="primary")
+        with gr.Column(scale=1):
+            emptyBtn = gr.Button("Clear History")
+            max_length = gr.Slider(0, 32768, value=8192, step=1.0, label="Maximum length", interactive=True)
+            top_p = gr.Slider(0, 1, value=0.8, step=0.01, label="Top P", interactive=True)
+            temperature = gr.Slider(0, 1, value=0.6, step=0.01, label="Temperature", interactive=True)
+
+    history = gr.State([])
+    past_key_values = gr.State(None)
+
+    submitBtn.click(predict, [user_input, chatbot, max_length, top_p, temperature, history, past_key_values],
+                    [chatbot, history, past_key_values], show_progress=True)
+    submitBtn.click(reset_user_input, [], [user_input])
+
+    emptyBtn.click(reset_state, outputs=[chatbot, history, past_key_values], show_progress=True)
+
+demo.queue().launch(share=False, server_name="127.0.0.1", server_port=8501, inbrowser=True)
diff --git a/basic_demo/web_demo2.py b/basic_demo/web_demo2.py
new file mode 100644
index 0000000000000000000000000000000000000000..e5e0e1c604c490e1bfbe30e88676d6d9c45733e9
--- /dev/null
+++ b/basic_demo/web_demo2.py
@@ -0,0 +1,89 @@
+import os
+import streamlit as st
+import torch
+from transformers import AutoModel, AutoTokenizer
+
+MODEL_PATH = os.environ.get('MODEL_PATH', 'THUDM/chatglm3-6b')
+TOKENIZER_PATH = os.environ.get("TOKENIZER_PATH", MODEL_PATH)
+DEVICE = 'cuda' if torch.cuda.is_available() else 'cpu'
+# 设置页面标题、图标和布局
+st.set_page_config(
+    page_title="ChatGLM3-6B 演示",
+    page_icon=":robot:",
+    layout="wide"
+)
+
+@st.cache_resource
+def get_model():
+    tokenizer = AutoTokenizer.from_pretrained(TOKENIZER_PATH, trust_remote_code=True)
+    if 'cuda' in DEVICE:  # AMD, NVIDIA GPU can use Half Precision
+        model = AutoModel.from_pretrained(MODEL_PATH, trust_remote_code=True).to(DEVICE).eval()
+    else:  # CPU, Intel GPU and other GPU can use Float16 Precision Only
+        model = AutoModel.from_pretrained(MODEL_PATH, trust_remote_code=True).float().to(DEVICE).eval()
+    # 多显卡支持,使用下面两行代替上面一行,将num_gpus改为你实际的显卡数量
+    # from utils import load_model_on_gpus
+    # model = load_model_on_gpus("THUDM/chatglm3-6b", num_gpus=2)
+    return tokenizer, model
+
+# 加载Chatglm3的model和tokenizer
+tokenizer, model = get_model()
+
+# 初始化历史记录和past key values
+if "history" not in st.session_state:
+    st.session_state.history = []
+if "past_key_values" not in st.session_state:
+    st.session_state.past_key_values = None
+
+# 设置max_length、top_p和temperature
+max_length = st.sidebar.slider("max_length", 0, 32768, 8192, step=1)
+top_p = st.sidebar.slider("top_p", 0.0, 1.0, 0.8, step=0.01)
+temperature = st.sidebar.slider("temperature", 0.0, 1.0, 0.6, step=0.01)
+
+# 清理会话历史
+buttonClean = st.sidebar.button("清理会话历史", key="clean")
+if buttonClean:
+    st.session_state.history = []
+    st.session_state.past_key_values = None
+    if torch.cuda.is_available():
+        torch.cuda.empty_cache()
+    st.rerun()
+
+# 渲染聊天历史记录
+for i, message in enumerate(st.session_state.history):
+    if message["role"] == "user":
+        with st.chat_message(name="user", avatar="user"):
+            st.markdown(message["content"])
+    else:
+        with st.chat_message(name="assistant", avatar="assistant"):
+            st.markdown(message["content"])
+
+# 输入框和输出框
+with st.chat_message(name="user", avatar="user"):
+    input_placeholder = st.empty()
+with st.chat_message(name="assistant", avatar="assistant"):
+    message_placeholder = st.empty()
+
+# 获取用户输入
+prompt_text = st.chat_input("请输入您的问题")
+
+# 如果用户输入了内容,则生成回复
+if prompt_text:
+
+    input_placeholder.markdown(prompt_text)
+    history = st.session_state.history
+    past_key_values = st.session_state.past_key_values
+    for response, history, past_key_values in model.stream_chat(
+        tokenizer,
+        prompt_text,
+        history,
+        past_key_values=past_key_values,
+        max_length=max_length,
+        top_p=top_p,
+        temperature=temperature,
+        return_past_key_values=True,
+    ):
+        message_placeholder.markdown(response)
+
+    # 更新历史记录和past key values
+    st.session_state.history = history
+    st.session_state.past_key_values = past_key_values
diff --git a/chatglm3-6b/.gitattributes b/chatglm3-6b/.gitattributes
new file mode 100644
index 0000000000000000000000000000000000000000..a6344aac8c09253b3b630fb776ae94478aa0275b
--- /dev/null
+++ b/chatglm3-6b/.gitattributes
@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
diff --git a/chatglm3-6b/MODEL_LICENSE b/chatglm3-6b/MODEL_LICENSE
new file mode 100644
index 0000000000000000000000000000000000000000..70ddc809f396cd1d05827a86d159fb72553ff5a9
--- /dev/null
+++ b/chatglm3-6b/MODEL_LICENSE
@@ -0,0 +1,65 @@
+The ChatGLM3-6B License
+
+1. 定义
+
+“许可方”是指分发其软件的 ChatGLM3-6B 模型团队。
+
+“软件”是指根据本许可提供的 ChatGLM3-6B 模型参数。
+
+2. 许可授予
+
+根据本许可的条款和条件，许可方特此授予您非排他性、全球性、不可转让、不可再许可、可撤销、免版税的版权许可。
+
+上述版权声明和本许可声明应包含在本软件的所有副本或重要部分中。
+
+3.限制
+
+您不得出于任何军事或非法目的使用、复制、修改、合并、发布、分发、复制或创建本软件的全部或部分衍生作品。
+
+您不得利用本软件从事任何危害国家安全和国家统一、危害社会公共利益、侵犯人身权益的行为。
+
+4.免责声明
+
+本软件“按原样”提供，不提供任何明示或暗示的保证，包括但不限于对适销性、特定用途的适用性和非侵权性的保证。 在任何情况下，作者或版权持有人均不对任何索赔、损害或其他责任负责，无论是在合同诉讼、侵权行为还是其他方面，由软件或软件的使用或其他交易引起、由软件引起或与之相关 软件。
+
+5. 责任限制
+
+除适用法律禁止的范围外，在任何情况下且根据任何法律理论，无论是基于侵权行为、疏忽、合同、责任或其他原因，任何许可方均不对您承担任何直接、间接、特殊、偶然、示范性、 或间接损害，或任何其他商业损失，即使许可人已被告知此类损害的可能性。
+
+6.争议解决
+
+本许可受中华人民共和国法律管辖并按其解释。 因本许可引起的或与本许可有关的任何争议应提交北京市海淀区人民法院。
+
+请注意，许可证可能会更新到更全面的版本。 有关许可和版权的任何问题，请通过 license@zhipuai.cn 与我们联系。
+
+1. Definitions
+
+“Licensor” means the ChatGLM3-6B Model Team that distributes its Software.
+
+“Software” means the ChatGLM3-6B model parameters made available under this license.
+
+2. License Grant
+
+Subject to the terms and conditions of this License, the Licensor hereby grants to you a non-exclusive, worldwide, non-transferable, non-sublicensable, revocable, royalty-free copyright license to use the Software.
+
+The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
+
+3. Restriction
+
+You will not use, copy, modify, merge, publish, distribute, reproduce, or create derivative works of the Software, in whole or in part, for any military, or illegal purposes.
+
+You will not use the Software for any act that may undermine China's national security and national unity, harm the public interest of society, or infringe upon the rights and interests of human beings.
+
+4. Disclaimer
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+
+5. Limitation of Liability
+
+EXCEPT TO THE EXTENT PROHIBITED BY APPLICABLE LAW, IN NO EVENT AND UNDER NO LEGAL THEORY, WHETHER BASED IN TORT, NEGLIGENCE, CONTRACT, LIABILITY, OR OTHERWISE WILL ANY LICENSOR BE LIABLE TO YOU FOR ANY DIRECT, INDIRECT, SPECIAL, INCIDENTAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES, OR ANY OTHER COMMERCIAL LOSSES, EVEN IF THE LICENSOR HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
+
+6. Dispute Resolution
+
+This license shall be governed and construed in accordance with the laws of People’s Republic of China. Any dispute arising from or in connection with this License shall be submitted to Haidian District People's Court in Beijing.
+
+Note that the license is subject to update to a more comprehensive version.  For any questions related to the license and copyright, please contact us at license@zhipuai.cn.
diff --git a/chatglm3-6b/README.md b/chatglm3-6b/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..7d7189ada4429ef969537867a7cff9c902f29d59
--- /dev/null
+++ b/chatglm3-6b/README.md
@@ -0,0 +1,102 @@
+---
+language:
+- zh
+- en
+tags:
+- glm
+- chatglm
+- thudm
+---
+# ChatGLM3-6B
+<p align="center">
+  💻 <a href="https://github.com/THUDM/ChatGLM" target="_blank">Github Repo</a> • 🐦 <a href="https://twitter.com/thukeg" target="_blank">Twitter</a> • 📃 <a href="https://arxiv.org/abs/2103.10360" target="_blank">[GLM@ACL 22]</a> <a href="https://github.com/THUDM/GLM" target="_blank">[GitHub]</a> • 📃 <a href="https://arxiv.org/abs/2210.02414" target="_blank">[GLM-130B@ICLR 23]</a> <a href="https://github.com/THUDM/GLM-130B" target="_blank">[GitHub]</a> <br>
+</p>
+
+<p align="center">
+    👋 Join our <a href="https://join.slack.com/t/chatglm/shared_invite/zt-25ti5uohv-A_hs~am_D3Q8XPZMpj7wwQ" target="_blank">Slack</a> and <a href="https://github.com/THUDM/ChatGLM/blob/main/resources/WECHAT.md" target="_blank">WeChat</a>
+</p>
+<p align="center">
+📍Experience the larger-scale ChatGLM model at <a href="https://www.chatglm.cn">chatglm.cn</a>
+</p>
+
+## 介绍 (Introduction)
+ChatGLM3-6B 是 ChatGLM 系列最新一代的开源模型，在保留了前两代模型对话流畅、部署门槛低等众多优秀特性的基础上，ChatGLM3-6B 引入了如下特性：
+
+1. **更强大的基础模型：** ChatGLM3-6B 的基础模型 ChatGLM3-6B-Base 采用了更多样的训练数据、更充分的训练步数和更合理的训练策略。在语义、数学、推理、代码、知识等不同角度的数据集上测评显示，ChatGLM3-6B-Base 具有在 10B 以下的预训练模型中最强的性能。
+2. **更完整的功能支持：** ChatGLM3-6B 采用了全新设计的 [Prompt 格式](https://github.com/THUDM/ChatGLM3/blob/main/PROMPT.md)，除正常的多轮对话外。同时原生支持[工具调用](https://github.com/THUDM/ChatGLM3/blob/main/tool_using/README.md)（Function Call）、代码执行（Code Interpreter）和 Agent 任务等复杂场景。
+3. **更全面的开源序列：** 除了对话模型 ChatGLM3-6B 外，还开源了基础模型 ChatGLM-6B-Base、长文本对话模型 ChatGLM3-6B-32K。以上所有权重对学术研究**完全开放**，在填写[问卷](https://open.bigmodel.cn/mla/form)进行登记后**亦允许免费商业使用**。
+
+ChatGLM3-6B is the latest open-source model in the ChatGLM series. While retaining many excellent features such as smooth dialogue and low deployment threshold from the previous two generations, ChatGLM3-6B introduces the following features:
+
+1. **More Powerful Base Model:** The base model of ChatGLM3-6B, ChatGLM3-6B-Base, employs a more diverse training dataset, more sufficient training steps, and a more reasonable training strategy. Evaluations on datasets such as semantics, mathematics, reasoning, code, knowledge, etc., show that ChatGLM3-6B-Base has the strongest performance among pre-trained models under 10B.
+2. **More Comprehensive Function Support:** ChatGLM3-6B adopts a newly designed [Prompt format](https://github.com/THUDM/ChatGLM3/blob/main/PROMPT_en.md), in addition to the normal multi-turn dialogue. It also natively supports [function call](https://github.com/THUDM/ChatGLM3/blob/main/tool_using/README_en.md), code interpreter, and complex scenarios such as agent tasks.
+3. **More Comprehensive Open-source Series:** In addition to the dialogue model ChatGLM3-6B, the base model ChatGLM-6B-Base and the long-text dialogue model ChatGLM3-6B-32K are also open-sourced. All the weights are **fully open** for academic research, and after completing the [questionnaire](https://open.bigmodel.cn/mla/form) registration, they are also **allowed for free commercial use**.
+
+## 软件依赖 (Dependencies)
+
+```shell
+pip install protobuf transformers==4.30.2 cpm_kernels torch>=2.0 gradio mdtex2html sentencepiece accelerate
+```
+
+## 代码调用 (Code Usage)
+
+可以通过如下代码调用 ChatGLM3-6B 模型来生成对话：
+
+You can generate dialogue by invoking the ChatGLM3-6B model with the following code:
+
+```ipython
+>>> from transformers import AutoTokenizer, AutoModel
+>>> tokenizer = AutoTokenizer.from_pretrained("THUDM/chatglm3-6b", trust_remote_code=True)
+>>> model = AutoModel.from_pretrained("THUDM/chatglm3-6b", trust_remote_code=True).half().cuda()
+>>> model = model.eval()
+>>> response, history = model.chat(tokenizer, "你好", history=[])
+>>> print(response)
+你好👋!我是人工智能助手 ChatGLM-6B,很高兴见到你,欢迎问我任何问题。
+>>> response, history = model.chat(tokenizer, "晚上睡不着应该怎么办", history=history)
+>>> print(response)
+晚上睡不着可能会让你感到焦虑或不舒服,但以下是一些可以帮助你入睡的方法:
+
+1. 制定规律的睡眠时间表:保持规律的睡眠时间表可以帮助你建立健康的睡眠习惯,使你更容易入睡。尽量在每天的相同时间上床,并在同一时间起床。
+2. 创造一个舒适的睡眠环境:确保睡眠环境舒适,安静,黑暗且温度适宜。可以使用舒适的床上用品,并保持房间通风。
+3. 放松身心:在睡前做些放松的活动,例如泡个热水澡,听些轻柔的音乐,阅读一些有趣的书籍等,有助于缓解紧张和焦虑,使你更容易入睡。
+4. 避免饮用含有咖啡因的饮料:咖啡因是一种刺激性物质,会影响你的睡眠质量。尽量避免在睡前饮用含有咖啡因的饮料,例如咖啡,茶和可乐。
+5. 避免在床上做与睡眠无关的事情:在床上做些与睡眠无关的事情,例如看电影,玩游戏或工作等,可能会干扰你的睡眠。
+6. 尝试呼吸技巧:深呼吸是一种放松技巧,可以帮助你缓解紧张和焦虑,使你更容易入睡。试着慢慢吸气,保持几秒钟,然后缓慢呼气。
+
+如果这些方法无法帮助你入睡,你可以考虑咨询医生或睡眠专家,寻求进一步的建议。
+```
+
+关于更多的使用说明，包括如何运行命令行和网页版本的 DEMO，以及使用模型量化以节省显存，请参考我们的 [Github Repo](https://github.com/THUDM/ChatGLM)。
+
+For more instructions, including how to run CLI and web demos, and model quantization, please refer to our [Github Repo](https://github.com/THUDM/ChatGLM).
+
+
+## 协议 (License)
+
+本仓库的代码依照 [Apache-2.0](LICENSE) 协议开源，ChatGLM3-6B 模型的权重的使用则需要遵循 [Model License](MODEL_LICENSE)。
+
+The code in this repository is open-sourced under the [Apache-2.0 license](LICENSE), while the use of the ChatGLM3-6B model weights needs to comply with the [Model License](MODEL_LICENSE).
+
+## 引用 (Citation)
+
+如果你觉得我们的工作有帮助的话，请考虑引用下列论文。
+
+If you find our work helpful, please consider citing the following papers.
+
+```
+@article{zeng2022glm,
+  title={Glm-130b: An open bilingual pre-trained model},
+  author={Zeng, Aohan and Liu, Xiao and Du, Zhengxiao and Wang, Zihan and Lai, Hanyu and Ding, Ming and Yang, Zhuoyi and Xu, Yifan and Zheng, Wendi and Xia, Xiao and others},
+  journal={arXiv preprint arXiv:2210.02414},
+  year={2022}
+}
+```
+```
+@inproceedings{du2022glm,
+  title={GLM: General Language Model Pretraining with Autoregressive Blank Infilling},
+  author={Du, Zhengxiao and Qian, Yujie and Liu, Xiao and Ding, Ming and Qiu, Jiezhong and Yang, Zhilin and Tang, Jie},
+  booktitle={Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
+  pages={320--335},
+  year={2022}
+}
+```
\ No newline at end of file
diff --git a/chatglm3-6b/config.json b/chatglm3-6b/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..37933c87f711fa0815a7712880bbf83eac4299c5
--- /dev/null
+++ b/chatglm3-6b/config.json
@@ -0,0 +1,42 @@
+{
+  "_name_or_path": "THUDM/chatglm3-6b",
+  "model_type": "chatglm",
+  "architectures": [
+    "ChatGLMModel"
+  ],
+  "auto_map": {
+    "AutoConfig": "configuration_chatglm.ChatGLMConfig",
+    "AutoModel": "modeling_chatglm.ChatGLMForConditionalGeneration",
+    "AutoModelForCausalLM": "modeling_chatglm.ChatGLMForConditionalGeneration",
+    "AutoModelForSeq2SeqLM": "modeling_chatglm.ChatGLMForConditionalGeneration",
+    "AutoModelForSequenceClassification": "modeling_chatglm.ChatGLMForSequenceClassification"
+  },
+  "add_bias_linear": false,
+  "add_qkv_bias": true,
+  "apply_query_key_layer_scaling": true,
+  "apply_residual_connection_post_layernorm": false,
+  "attention_dropout": 0.0,
+  "attention_softmax_in_fp32": true,
+  "bias_dropout_fusion": true,
+  "ffn_hidden_size": 13696,
+  "fp32_residual_connection": false,
+  "hidden_dropout": 0.0,
+  "hidden_size": 4096,
+  "kv_channels": 128,
+  "layernorm_epsilon": 1e-05,
+  "multi_query_attention": true,
+  "multi_query_group_num": 2,
+  "num_attention_heads": 32,
+  "num_layers": 28,
+  "original_rope": true,
+  "padded_vocab_size": 65024,
+  "post_layer_norm": true,
+  "rmsnorm": true,
+  "seq_length": 8192,
+  "use_cache": true,
+  "torch_dtype": "float16",
+  "transformers_version": "4.30.2",
+  "tie_word_embeddings": false,
+  "eos_token_id": 2,
+  "pad_token_id": 0
+}
\ No newline at end of file
diff --git a/chatglm3-6b/configuration_chatglm.py b/chatglm3-6b/configuration_chatglm.py
new file mode 100644
index 0000000000000000000000000000000000000000..35600185f5a26951081de0f3a41a913eaf06af99
--- /dev/null
+++ b/chatglm3-6b/configuration_chatglm.py
@@ -0,0 +1,61 @@
+from transformers import PretrainedConfig
+
+
+class ChatGLMConfig(PretrainedConfig):
+    model_type = "chatglm"
+    def __init__(
+        self,
+        num_layers=28,
+        padded_vocab_size=65024,
+        hidden_size=4096,
+        ffn_hidden_size=13696,
+        kv_channels=128,
+        num_attention_heads=32,
+        seq_length=2048,
+        hidden_dropout=0.0,
+        classifier_dropout=None,
+        attention_dropout=0.0,
+        layernorm_epsilon=1e-5,
+        rmsnorm=True,
+        apply_residual_connection_post_layernorm=False,
+        post_layer_norm=True,
+        add_bias_linear=False,
+        add_qkv_bias=False,
+        bias_dropout_fusion=True,
+        multi_query_attention=False,
+        multi_query_group_num=1,
+        apply_query_key_layer_scaling=True,
+        attention_softmax_in_fp32=True,
+        fp32_residual_connection=False,
+        quantization_bit=0,
+        pre_seq_len=None,
+        prefix_projection=False,
+        **kwargs
+    ):
+        self.num_layers = num_layers
+        self.vocab_size = padded_vocab_size
+        self.padded_vocab_size = padded_vocab_size
+        self.hidden_size = hidden_size
+        self.ffn_hidden_size = ffn_hidden_size
+        self.kv_channels = kv_channels
+        self.num_attention_heads = num_attention_heads
+        self.seq_length = seq_length
+        self.hidden_dropout = hidden_dropout
+        self.classifier_dropout = classifier_dropout
+        self.attention_dropout = attention_dropout
+        self.layernorm_epsilon = layernorm_epsilon
+        self.rmsnorm = rmsnorm
+        self.apply_residual_connection_post_layernorm = apply_residual_connection_post_layernorm
+        self.post_layer_norm = post_layer_norm
+        self.add_bias_linear = add_bias_linear
+        self.add_qkv_bias = add_qkv_bias
+        self.bias_dropout_fusion = bias_dropout_fusion
+        self.multi_query_attention = multi_query_attention
+        self.multi_query_group_num = multi_query_group_num
+        self.apply_query_key_layer_scaling = apply_query_key_layer_scaling
+        self.attention_softmax_in_fp32 = attention_softmax_in_fp32
+        self.fp32_residual_connection = fp32_residual_connection
+        self.quantization_bit = quantization_bit
+        self.pre_seq_len = pre_seq_len
+        self.prefix_projection = prefix_projection
+        super().__init__(**kwargs)
\ No newline at end of file
diff --git a/chatglm3-6b/modeling_chatglm.py b/chatglm3-6b/modeling_chatglm.py
new file mode 100644
index 0000000000000000000000000000000000000000..c5b5027587016090a377f25289284b6e4f829cb4
--- /dev/null
+++ b/chatglm3-6b/modeling_chatglm.py
@@ -0,0 +1,1293 @@
+""" PyTorch ChatGLM model. """
+
+import math
+import copy
+import warnings
+import re
+import sys
+
+import torch
+import torch.utils.checkpoint
+import torch.nn.functional as F
+from torch import nn
+from torch.nn import CrossEntropyLoss, LayerNorm, MSELoss, BCEWithLogitsLoss
+from torch.nn.utils import skip_init
+from typing import Optional, Tuple, Union, List, Callable, Dict, Any
+from copy import deepcopy
+
+from transformers.modeling_outputs import (
+    BaseModelOutputWithPast,
+    CausalLMOutputWithPast,
+    SequenceClassifierOutputWithPast,
+)
+from transformers.modeling_utils import PreTrainedModel
+from transformers.utils import logging
+from transformers.generation.logits_process import LogitsProcessor
+from transformers.generation.utils import LogitsProcessorList, StoppingCriteriaList, GenerationConfig, ModelOutput
+
+from .configuration_chatglm import ChatGLMConfig
+
+# flags required to enable jit fusion kernels
+
+if sys.platform != 'darwin':
+    torch._C._jit_set_profiling_mode(False)
+    torch._C._jit_set_profiling_executor(False)
+    torch._C._jit_override_can_fuse_on_cpu(True)
+    torch._C._jit_override_can_fuse_on_gpu(True)
+
+logger = logging.get_logger(__name__)
+
+_CHECKPOINT_FOR_DOC = "THUDM/ChatGLM"
+_CONFIG_FOR_DOC = "ChatGLMConfig"
+
+CHATGLM_6B_PRETRAINED_MODEL_ARCHIVE_LIST = [
+    "THUDM/chatglm3-6b",
+    # See all ChatGLM models at https://huggingface.co/models?filter=chatglm
+]
+
+
+def default_init(cls, *args, **kwargs):
+    return cls(*args, **kwargs)
+
+
+class InvalidScoreLogitsProcessor(LogitsProcessor):
+    def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTensor) -> torch.FloatTensor:
+        if torch.isnan(scores).any() or torch.isinf(scores).any():
+            scores.zero_()
+            scores[..., 5] = 5e4
+        return scores
+
+
+class PrefixEncoder(torch.nn.Module):
+    """
+    The torch.nn model to encode the prefix
+    Input shape: (batch-size, prefix-length)
+    Output shape: (batch-size, prefix-length, 2*layers*hidden)
+    """
+
+    def __init__(self, config: ChatGLMConfig):
+        super().__init__()
+        self.prefix_projection = config.prefix_projection
+        if self.prefix_projection:
+            # Use a two-layer MLP to encode the prefix
+            kv_size = config.num_layers * config.kv_channels * config.multi_query_group_num * 2
+            self.embedding = torch.nn.Embedding(config.pre_seq_len, kv_size)
+            self.trans = torch.nn.Sequential(
+                torch.nn.Linear(kv_size, config.hidden_size),
+                torch.nn.Tanh(),
+                torch.nn.Linear(config.hidden_size, kv_size)
+            )
+        else:
+            self.embedding = torch.nn.Embedding(config.pre_seq_len,
+                                                config.num_layers * config.kv_channels * config.multi_query_group_num * 2)
+
+    def forward(self, prefix: torch.Tensor):
+        if self.prefix_projection:
+            prefix_tokens = self.embedding(prefix)
+            past_key_values = self.trans(prefix_tokens)
+        else:
+            past_key_values = self.embedding(prefix)
+        return past_key_values
+
+
+def split_tensor_along_last_dim(
+        tensor: torch.Tensor,
+        num_partitions: int,
+        contiguous_split_chunks: bool = False,
+) -> List[torch.Tensor]:
+    """Split a tensor along its last dimension.
+
+    Arguments:
+        tensor: input tensor.
+        num_partitions: number of partitions to split the tensor
+        contiguous_split_chunks: If True, make each chunk contiguous
+                                 in memory.
+
+    Returns:
+        A list of Tensors
+    """
+    # Get the size and dimension.
+    last_dim = tensor.dim() - 1
+    last_dim_size = tensor.size()[last_dim] // num_partitions
+    # Split.
+    tensor_list = torch.split(tensor, last_dim_size, dim=last_dim)
+    # Note: torch.split does not create contiguous tensors by default.
+    if contiguous_split_chunks:
+        return tuple(chunk.contiguous() for chunk in tensor_list)
+
+    return tensor_list
+
+
+class RotaryEmbedding(nn.Module):
+    def __init__(self, dim, original_impl=False, device=None, dtype=None):
+        super().__init__()
+        inv_freq = 1.0 / (10000 ** (torch.arange(0, dim, 2, device=device).to(dtype=dtype) / dim))
+        self.register_buffer("inv_freq", inv_freq)
+        self.dim = dim
+        self.original_impl = original_impl
+
+    def forward_impl(
+            self, seq_len: int, n_elem: int, dtype: torch.dtype, device: torch.device, base: int = 10000
+    ):
+        """Enhanced Transformer with Rotary Position Embedding.
+
+        Derived from: https://github.com/labmlai/annotated_deep_learning_paper_implementations/blob/master/labml_nn/
+        transformers/rope/__init__.py. MIT License:
+        https://github.com/labmlai/annotated_deep_learning_paper_implementations/blob/master/license.
+        """
+        # $\Theta = {\theta_i = 10000^{\frac{2(i-1)}{d}}, i \in [1, 2, ..., \frac{d}{2}]}$
+        theta = 1.0 / (base ** (torch.arange(0, n_elem, 2, dtype=torch.float, device=device) / n_elem))
+
+        # Create position indexes `[0, 1, ..., seq_len - 1]`
+        seq_idx = torch.arange(seq_len, dtype=torch.float, device=device)
+
+        # Calculate the product of position index and $\theta_i$
+        idx_theta = torch.outer(seq_idx, theta).float()
+
+        cache = torch.stack([torch.cos(idx_theta), torch.sin(idx_theta)], dim=-1)
+
+        # this is to mimic the behaviour of complex32, else we will get different results
+        if dtype in (torch.float16, torch.bfloat16, torch.int8):
+            cache = cache.bfloat16() if dtype == torch.bfloat16 else cache.half()
+        return cache
+
+    def forward(self, max_seq_len, offset=0):
+        return self.forward_impl(
+            max_seq_len, self.dim, dtype=self.inv_freq.dtype, device=self.inv_freq.device
+        )
+
+
+@torch.jit.script
+def apply_rotary_pos_emb(x: torch.Tensor, rope_cache: torch.Tensor) -> torch.Tensor:
+    # x: [sq, b, np, hn]
+    sq, b, np, hn = x.size(0), x.size(1), x.size(2), x.size(3)
+    rot_dim = rope_cache.shape[-2] * 2
+    x, x_pass = x[..., :rot_dim], x[..., rot_dim:]
+    # truncate to support variable sizes
+    rope_cache = rope_cache[:sq]
+    xshaped = x.reshape(sq, -1, np, rot_dim // 2, 2)
+    rope_cache = rope_cache.view(sq, -1, 1, xshaped.size(3), 2)
+    x_out2 = torch.stack(
+        [
+            xshaped[..., 0] * rope_cache[..., 0] - xshaped[..., 1] * rope_cache[..., 1],
+            xshaped[..., 1] * rope_cache[..., 0] + xshaped[..., 0] * rope_cache[..., 1],
+        ],
+        -1,
+    )
+    x_out2 = x_out2.flatten(3)
+    return torch.cat((x_out2, x_pass), dim=-1)
+
+
+class RMSNorm(torch.nn.Module):
+    def __init__(self, normalized_shape, eps=1e-5, device=None, dtype=None, **kwargs):
+        super().__init__()
+        self.weight = torch.nn.Parameter(torch.empty(normalized_shape, device=device, dtype=dtype))
+        self.eps = eps
+
+    def forward(self, hidden_states: torch.Tensor):
+        input_dtype = hidden_states.dtype
+        variance = hidden_states.to(torch.float32).pow(2).mean(-1, keepdim=True)
+        hidden_states = hidden_states * torch.rsqrt(variance + self.eps)
+
+        return (self.weight * hidden_states).to(input_dtype)
+
+
+class CoreAttention(torch.nn.Module):
+    def __init__(self, config: ChatGLMConfig, layer_number):
+        super(CoreAttention, self).__init__()
+
+        self.apply_query_key_layer_scaling = config.apply_query_key_layer_scaling
+        self.attention_softmax_in_fp32 = config.attention_softmax_in_fp32
+        if self.apply_query_key_layer_scaling:
+            self.attention_softmax_in_fp32 = True
+        self.layer_number = max(1, layer_number)
+
+        projection_size = config.kv_channels * config.num_attention_heads
+
+        # Per attention head and per partition values.
+        self.hidden_size_per_partition = projection_size
+        self.hidden_size_per_attention_head = projection_size // config.num_attention_heads
+        self.num_attention_heads_per_partition = config.num_attention_heads
+
+        coeff = None
+        self.norm_factor = math.sqrt(self.hidden_size_per_attention_head)
+        if self.apply_query_key_layer_scaling:
+            coeff = self.layer_number
+            self.norm_factor *= coeff
+        self.coeff = coeff
+
+        self.attention_dropout = torch.nn.Dropout(config.attention_dropout)
+
+    def forward(self, query_layer, key_layer, value_layer, attention_mask):
+        pytorch_major_version = int(torch.__version__.split('.')[0])
+        if pytorch_major_version >= 2:
+            query_layer, key_layer, value_layer = [k.permute(1, 2, 0, 3) for k in [query_layer, key_layer, value_layer]]
+            if attention_mask is None and query_layer.shape[2] == key_layer.shape[2]:
+                context_layer = torch.nn.functional.scaled_dot_product_attention(query_layer, key_layer, value_layer,
+                                                                                 is_causal=True)
+            else:
+                if attention_mask is not None:
+                    attention_mask = ~attention_mask
+                context_layer = torch.nn.functional.scaled_dot_product_attention(query_layer, key_layer, value_layer,
+                                                                                 attention_mask)
+            context_layer = context_layer.permute(2, 0, 1, 3)
+            new_context_layer_shape = context_layer.size()[:-2] + (self.hidden_size_per_partition,)
+            context_layer = context_layer.reshape(*new_context_layer_shape)
+        else:
+            # Raw attention scores
+
+            # [b, np, sq, sk]
+            output_size = (query_layer.size(1), query_layer.size(2), query_layer.size(0), key_layer.size(0))
+
+            # [sq, b, np, hn] -> [sq, b * np, hn]
+            query_layer = query_layer.view(output_size[2], output_size[0] * output_size[1], -1)
+            # [sk, b, np, hn] -> [sk, b * np, hn]
+            key_layer = key_layer.view(output_size[3], output_size[0] * output_size[1], -1)
+
+            # preallocting input tensor: [b * np, sq, sk]
+            matmul_input_buffer = torch.empty(
+                output_size[0] * output_size[1], output_size[2], output_size[3], dtype=query_layer.dtype,
+                device=query_layer.device
+            )
+
+            # Raw attention scores. [b * np, sq, sk]
+            matmul_result = torch.baddbmm(
+                matmul_input_buffer,
+                query_layer.transpose(0, 1),  # [b * np, sq, hn]
+                key_layer.transpose(0, 1).transpose(1, 2),  # [b * np, hn, sk]
+                beta=0.0,
+                alpha=(1.0 / self.norm_factor),
+            )
+
+            # change view to [b, np, sq, sk]
+            attention_scores = matmul_result.view(*output_size)
+
+            # ===========================
+            # Attention probs and dropout
+            # ===========================
+
+            # attention scores and attention mask [b, np, sq, sk]
+            if self.attention_softmax_in_fp32:
+                attention_scores = attention_scores.float()
+            if self.coeff is not None:
+                attention_scores = attention_scores * self.coeff
+            if attention_mask is None and attention_scores.shape[2] == attention_scores.shape[3]:
+                attention_mask = torch.ones(output_size[0], 1, output_size[2], output_size[3],
+                                            device=attention_scores.device, dtype=torch.bool)
+                attention_mask.tril_()
+                attention_mask = ~attention_mask
+            if attention_mask is not None:
+                attention_scores = attention_scores.masked_fill(attention_mask, float("-inf"))
+            attention_probs = F.softmax(attention_scores, dim=-1)
+            attention_probs = attention_probs.type_as(value_layer)
+
+            # This is actually dropping out entire tokens to attend to, which might
+            # seem a bit unusual, but is taken from the original Transformer paper.
+            attention_probs = self.attention_dropout(attention_probs)
+            # =========================
+            # Context layer. [sq, b, hp]
+            # =========================
+
+            # value_layer -> context layer.
+            # [sk, b, np, hn] --> [b, np, sq, hn]
+
+            # context layer shape: [b, np, sq, hn]
+            output_size = (value_layer.size(1), value_layer.size(2), query_layer.size(0), value_layer.size(3))
+            # change view [sk, b * np, hn]
+            value_layer = value_layer.view(value_layer.size(0), output_size[0] * output_size[1], -1)
+            # change view [b * np, sq, sk]
+            attention_probs = attention_probs.view(output_size[0] * output_size[1], output_size[2], -1)
+            # matmul: [b * np, sq, hn]
+            context_layer = torch.bmm(attention_probs, value_layer.transpose(0, 1))
+            # change view [b, np, sq, hn]
+            context_layer = context_layer.view(*output_size)
+            # [b, np, sq, hn] --> [sq, b, np, hn]
+            context_layer = context_layer.permute(2, 0, 1, 3).contiguous()
+            # [sq, b, np, hn] --> [sq, b, hp]
+            new_context_layer_shape = context_layer.size()[:-2] + (self.hidden_size_per_partition,)
+            context_layer = context_layer.view(*new_context_layer_shape)
+
+        return context_layer
+
+
+class SelfAttention(torch.nn.Module):
+    """Parallel self-attention layer abstract class.
+
+    Self-attention layer takes input with size [s, b, h]
+    and returns output of the same size.
+    """
+
+    def __init__(self, config: ChatGLMConfig, layer_number, device=None):
+        super(SelfAttention, self).__init__()
+        self.layer_number = max(1, layer_number)
+
+        self.projection_size = config.kv_channels * config.num_attention_heads
+
+        # Per attention head and per partition values.
+        self.hidden_size_per_attention_head = self.projection_size // config.num_attention_heads
+        self.num_attention_heads_per_partition = config.num_attention_heads
+
+        self.multi_query_attention = config.multi_query_attention
+        self.qkv_hidden_size = 3 * self.projection_size
+        if self.multi_query_attention:
+            self.num_multi_query_groups_per_partition = config.multi_query_group_num
+            self.qkv_hidden_size = (
+                    self.projection_size + 2 * self.hidden_size_per_attention_head * config.multi_query_group_num
+            )
+        self.query_key_value = nn.Linear(config.hidden_size, self.qkv_hidden_size,
+                                         bias=config.add_bias_linear or config.add_qkv_bias,
+                                         device=device, **_config_to_kwargs(config)
+                                         )
+
+        self.core_attention = CoreAttention(config, self.layer_number)
+
+        # Output.
+        self.dense = nn.Linear(self.projection_size, config.hidden_size, bias=config.add_bias_linear,
+                               device=device, **_config_to_kwargs(config)
+                               )
+
+    def _allocate_memory(self, inference_max_sequence_len, batch_size, device=None, dtype=None):
+        if self.multi_query_attention:
+            num_attention_heads = self.num_multi_query_groups_per_partition
+        else:
+            num_attention_heads = self.num_attention_heads_per_partition
+        return torch.empty(
+            inference_max_sequence_len,
+            batch_size,
+            num_attention_heads,
+            self.hidden_size_per_attention_head,
+            dtype=dtype,
+            device=device,
+        )
+
+    def forward(
+            self, hidden_states, attention_mask, rotary_pos_emb, kv_cache=None, use_cache=True
+    ):
+        # hidden_states: [sq, b, h]
+
+        # =================================================
+        # Pre-allocate memory for key-values for inference.
+        # =================================================
+        # =====================
+        # Query, Key, and Value
+        # =====================
+
+        # Attention heads [sq, b, h] --> [sq, b, (np * 3 * hn)]
+        mixed_x_layer = self.query_key_value(hidden_states)
+
+        if self.multi_query_attention:
+            (query_layer, key_layer, value_layer) = mixed_x_layer.split(
+                [
+                    self.num_attention_heads_per_partition * self.hidden_size_per_attention_head,
+                    self.num_multi_query_groups_per_partition * self.hidden_size_per_attention_head,
+                    self.num_multi_query_groups_per_partition * self.hidden_size_per_attention_head,
+                ],
+                dim=-1,
+            )
+            query_layer = query_layer.view(
+                query_layer.size()[:-1] + (self.num_attention_heads_per_partition, self.hidden_size_per_attention_head)
+            )
+            key_layer = key_layer.view(
+                key_layer.size()[:-1] + (self.num_multi_query_groups_per_partition, self.hidden_size_per_attention_head)
+            )
+            value_layer = value_layer.view(
+                value_layer.size()[:-1]
+                + (self.num_multi_query_groups_per_partition, self.hidden_size_per_attention_head)
+            )
+        else:
+            new_tensor_shape = mixed_x_layer.size()[:-1] + \
+                               (self.num_attention_heads_per_partition,
+                                3 * self.hidden_size_per_attention_head)
+            mixed_x_layer = mixed_x_layer.view(*new_tensor_shape)
+
+            # [sq, b, np, 3 * hn] --> 3 [sq, b, np, hn]
+            (query_layer, key_layer, value_layer) = split_tensor_along_last_dim(mixed_x_layer, 3)
+
+        # apply relative positional encoding (rotary embedding)
+        if rotary_pos_emb is not None:
+            query_layer = apply_rotary_pos_emb(query_layer, rotary_pos_emb)
+            key_layer = apply_rotary_pos_emb(key_layer, rotary_pos_emb)
+
+        # adjust key and value for inference
+        if kv_cache is not None:
+            cache_k, cache_v = kv_cache
+            key_layer = torch.cat((cache_k, key_layer), dim=0)
+            value_layer = torch.cat((cache_v, value_layer), dim=0)
+        if use_cache:
+            kv_cache = (key_layer, value_layer)
+        else:
+            kv_cache = None
+
+        if self.multi_query_attention:
+            key_layer = key_layer.unsqueeze(-2)
+            key_layer = key_layer.expand(
+                -1, -1, -1, self.num_attention_heads_per_partition // self.num_multi_query_groups_per_partition, -1
+            )
+            key_layer = key_layer.contiguous().view(
+                key_layer.size()[:2] + (self.num_attention_heads_per_partition, self.hidden_size_per_attention_head)
+            )
+            value_layer = value_layer.unsqueeze(-2)
+            value_layer = value_layer.expand(
+                -1, -1, -1, self.num_attention_heads_per_partition // self.num_multi_query_groups_per_partition, -1
+            )
+            value_layer = value_layer.contiguous().view(
+                value_layer.size()[:2] + (self.num_attention_heads_per_partition, self.hidden_size_per_attention_head)
+            )
+
+        # ==================================
+        # core attention computation
+        # ==================================
+
+        context_layer = self.core_attention(query_layer, key_layer, value_layer, attention_mask)
+
+        # =================
+        # Output. [sq, b, h]
+        # =================
+
+        output = self.dense(context_layer)
+
+        return output, kv_cache
+
+
+def _config_to_kwargs(args):
+    common_kwargs = {
+        "dtype": args.torch_dtype,
+    }
+    return common_kwargs
+
+
+class MLP(torch.nn.Module):
+    """MLP.
+
+    MLP will take the input with h hidden state, project it to 4*h
+    hidden dimension, perform nonlinear transformation, and project the
+    state back into h hidden dimension.
+    """
+
+    def __init__(self, config: ChatGLMConfig, device=None):
+        super(MLP, self).__init__()
+
+        self.add_bias = config.add_bias_linear
+
+        # Project to 4h. If using swiglu double the output width, see https://arxiv.org/pdf/2002.05202.pdf
+        self.dense_h_to_4h = nn.Linear(
+            config.hidden_size,
+            config.ffn_hidden_size * 2,
+            bias=self.add_bias,
+            device=device,
+            **_config_to_kwargs(config)
+        )
+
+        def swiglu(x):
+            x = torch.chunk(x, 2, dim=-1)
+            return F.silu(x[0]) * x[1]
+
+        self.activation_func = swiglu
+
+        # Project back to h.
+        self.dense_4h_to_h = nn.Linear(
+            config.ffn_hidden_size,
+            config.hidden_size,
+            bias=self.add_bias,
+            device=device,
+            **_config_to_kwargs(config)
+        )
+
+    def forward(self, hidden_states):
+        # [s, b, 4hp]
+        intermediate_parallel = self.dense_h_to_4h(hidden_states)
+        intermediate_parallel = self.activation_func(intermediate_parallel)
+        # [s, b, h]
+        output = self.dense_4h_to_h(intermediate_parallel)
+        return output
+
+
+class GLMBlock(torch.nn.Module):
+    """A single transformer layer.
+
+    Transformer layer takes input with size [s, b, h] and returns an
+    output of the same size.
+    """
+
+    def __init__(self, config: ChatGLMConfig, layer_number, device=None):
+        super(GLMBlock, self).__init__()
+        self.layer_number = layer_number
+
+        self.apply_residual_connection_post_layernorm = config.apply_residual_connection_post_layernorm
+
+        self.fp32_residual_connection = config.fp32_residual_connection
+
+        LayerNormFunc = RMSNorm if config.rmsnorm else LayerNorm
+        # Layernorm on the input data.
+        self.input_layernorm = LayerNormFunc(config.hidden_size, eps=config.layernorm_epsilon, device=device,
+                                             dtype=config.torch_dtype)
+
+        # Self attention.
+        self.self_attention = SelfAttention(config, layer_number, device=device)
+        self.hidden_dropout = config.hidden_dropout
+
+        # Layernorm on the attention output
+        self.post_attention_layernorm = LayerNormFunc(config.hidden_size, eps=config.layernorm_epsilon, device=device,
+                                                      dtype=config.torch_dtype)
+
+        # MLP
+        self.mlp = MLP(config, device=device)
+
+    def forward(
+            self, hidden_states, attention_mask, rotary_pos_emb, kv_cache=None, use_cache=True,
+    ):
+        # hidden_states: [s, b, h]
+
+        # Layer norm at the beginning of the transformer layer.
+        layernorm_output = self.input_layernorm(hidden_states)
+        # Self attention.
+        attention_output, kv_cache = self.self_attention(
+            layernorm_output,
+            attention_mask,
+            rotary_pos_emb,
+            kv_cache=kv_cache,
+            use_cache=use_cache
+        )
+
+        # Residual connection.
+        if self.apply_residual_connection_post_layernorm:
+            residual = layernorm_output
+        else:
+            residual = hidden_states
+
+        layernorm_input = torch.nn.functional.dropout(attention_output, p=self.hidden_dropout, training=self.training)
+        layernorm_input = residual + layernorm_input
+
+        # Layer norm post the self attention.
+        layernorm_output = self.post_attention_layernorm(layernorm_input)
+
+        # MLP.
+        mlp_output = self.mlp(layernorm_output)
+
+        # Second residual connection.
+        if self.apply_residual_connection_post_layernorm:
+            residual = layernorm_output
+        else:
+            residual = layernorm_input
+
+        output = torch.nn.functional.dropout(mlp_output, p=self.hidden_dropout, training=self.training)
+        output = residual + output
+
+        return output, kv_cache
+
+
+class GLMTransformer(torch.nn.Module):
+    """Transformer class."""
+
+    def __init__(self, config: ChatGLMConfig, device=None):
+        super(GLMTransformer, self).__init__()
+
+        self.fp32_residual_connection = config.fp32_residual_connection
+        self.post_layer_norm = config.post_layer_norm
+
+        # Number of layers.
+        self.num_layers = config.num_layers
+
+        # Transformer layers.
+        def build_layer(layer_number):
+            return GLMBlock(config, layer_number, device=device)
+
+        self.layers = torch.nn.ModuleList([build_layer(i + 1) for i in range(self.num_layers)])
+
+        if self.post_layer_norm:
+            LayerNormFunc = RMSNorm if config.rmsnorm else LayerNorm
+            # Final layer norm before output.
+            self.final_layernorm = LayerNormFunc(config.hidden_size, eps=config.layernorm_epsilon, device=device,
+                                                 dtype=config.torch_dtype)
+
+        self.gradient_checkpointing = False
+
+    def _get_layer(self, layer_number):
+        return self.layers[layer_number]
+
+    def forward(
+            self, hidden_states, attention_mask, rotary_pos_emb, kv_caches=None,
+            use_cache: Optional[bool] = True,
+            output_hidden_states: Optional[bool] = False,
+    ):
+        if not kv_caches:
+            kv_caches = [None for _ in range(self.num_layers)]
+        presents = () if use_cache else None
+        if self.gradient_checkpointing and self.training:
+            if use_cache:
+                logger.warning_once(
+                    "`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`..."
+                )
+                use_cache = False
+
+        all_self_attentions = None
+        all_hidden_states = () if output_hidden_states else None
+        for index in range(self.num_layers):
+            if output_hidden_states:
+                all_hidden_states = all_hidden_states + (hidden_states,)
+
+            layer = self._get_layer(index)
+            if self.gradient_checkpointing and self.training:
+                layer_ret = torch.utils.checkpoint.checkpoint(
+                    layer,
+                    hidden_states,
+                    attention_mask,
+                    rotary_pos_emb,
+                    kv_caches[index],
+                    use_cache
+                )
+            else:
+                layer_ret = layer(
+                    hidden_states,
+                    attention_mask,
+                    rotary_pos_emb,
+                    kv_cache=kv_caches[index],
+                    use_cache=use_cache
+                )
+            hidden_states, kv_cache = layer_ret
+            if use_cache:
+                presents = presents + (kv_cache,)
+
+        if output_hidden_states:
+            all_hidden_states = all_hidden_states + (hidden_states,)
+
+        # Final layer norm.
+        if self.post_layer_norm:
+            hidden_states = self.final_layernorm(hidden_states)
+
+        return hidden_states, presents, all_hidden_states, all_self_attentions
+
+
+class ChatGLMPreTrainedModel(PreTrainedModel):
+    """
+    An abstract class to handle weights initialization and
+    a simple interface for downloading and loading pretrained models.
+    """
+
+    is_parallelizable = False
+    supports_gradient_checkpointing = True
+    config_class = ChatGLMConfig
+    base_model_prefix = "transformer"
+    _no_split_modules = ["GLMBlock"]
+
+    def _init_weights(self, module: nn.Module):
+        """Initialize the weights."""
+        return
+
+    def get_masks(self, input_ids, past_key_values, padding_mask=None):
+        batch_size, seq_length = input_ids.shape
+        full_attention_mask = torch.ones(batch_size, seq_length, seq_length, device=input_ids.device)
+        full_attention_mask.tril_()
+        past_length = 0
+        if past_key_values:
+            past_length = past_key_values[0][0].shape[0]
+        if past_length:
+            full_attention_mask = torch.cat((torch.ones(batch_size, seq_length, past_length,
+                                                        device=input_ids.device), full_attention_mask), dim=-1)
+        if padding_mask is not None:
+            full_attention_mask = full_attention_mask * padding_mask.unsqueeze(1)
+        if not past_length and padding_mask is not None:
+            full_attention_mask -= padding_mask.unsqueeze(-1) - 1
+        full_attention_mask = (full_attention_mask < 0.5).bool()
+        full_attention_mask.unsqueeze_(1)
+        return full_attention_mask
+
+    def get_position_ids(self, input_ids, device):
+        batch_size, seq_length = input_ids.shape
+        position_ids = torch.arange(seq_length, dtype=torch.long, device=device).unsqueeze(0).repeat(batch_size, 1)
+        return position_ids
+
+    def _set_gradient_checkpointing(self, module, value=False):
+        if isinstance(module, GLMTransformer):
+            module.gradient_checkpointing = value
+
+
+class Embedding(torch.nn.Module):
+    """Language model embeddings."""
+
+    def __init__(self, config: ChatGLMConfig, device=None):
+        super(Embedding, self).__init__()
+
+        self.hidden_size = config.hidden_size
+        # Word embeddings (parallel).
+        self.word_embeddings = nn.Embedding(
+            config.padded_vocab_size,
+            self.hidden_size,
+            dtype=config.torch_dtype,
+            device=device
+        )
+        self.fp32_residual_connection = config.fp32_residual_connection
+
+    def forward(self, input_ids):
+        # Embeddings.
+        words_embeddings = self.word_embeddings(input_ids)
+        embeddings = words_embeddings
+        # Data format change to avoid explicit tranposes : [b s h] --> [s b h].
+        embeddings = embeddings.transpose(0, 1).contiguous()
+        # If the input flag for fp32 residual connection is set, convert for float.
+        if self.fp32_residual_connection:
+            embeddings = embeddings.float()
+        return embeddings
+
+
+class ChatGLMModel(ChatGLMPreTrainedModel):
+    def __init__(self, config: ChatGLMConfig, device=None, empty_init=True):
+        super().__init__(config)
+        if empty_init:
+            init_method = skip_init
+        else:
+            init_method = default_init
+        init_kwargs = {}
+        if device is not None:
+            init_kwargs["device"] = device
+        self.embedding = init_method(Embedding, config, **init_kwargs)
+        self.num_layers = config.num_layers
+        self.multi_query_group_num = config.multi_query_group_num
+        self.kv_channels = config.kv_channels
+
+        # Rotary positional embeddings
+        self.seq_length = config.seq_length
+        rotary_dim = (
+            config.hidden_size // config.num_attention_heads if config.kv_channels is None else config.kv_channels
+        )
+
+        self.rotary_pos_emb = RotaryEmbedding(rotary_dim // 2, original_impl=config.original_rope, device=device,
+                                              dtype=config.torch_dtype)
+        self.encoder = init_method(GLMTransformer, config, **init_kwargs)
+        self.output_layer = init_method(nn.Linear, config.hidden_size, config.padded_vocab_size, bias=False,
+                                        dtype=config.torch_dtype, **init_kwargs)
+        self.pre_seq_len = config.pre_seq_len
+        self.prefix_projection = config.prefix_projection
+        if self.pre_seq_len is not None:
+            for param in self.parameters():
+                param.requires_grad = False
+            self.prefix_tokens = torch.arange(self.pre_seq_len).long()
+            self.prefix_encoder = PrefixEncoder(config)
+            self.dropout = torch.nn.Dropout(0.1)
+
+    def get_input_embeddings(self):
+        return self.embedding.word_embeddings
+
+    def get_prompt(self, batch_size, device, dtype=torch.half):
+        prefix_tokens = self.prefix_tokens.unsqueeze(0).expand(batch_size, -1).to(device)
+        past_key_values = self.prefix_encoder(prefix_tokens).type(dtype)
+        past_key_values = past_key_values.view(
+            batch_size,
+            self.pre_seq_len,
+            self.num_layers * 2,
+            self.multi_query_group_num,
+            self.kv_channels
+        )
+        # seq_len, b, nh, hidden_size
+        past_key_values = self.dropout(past_key_values)
+        past_key_values = past_key_values.permute([2, 1, 0, 3, 4]).split(2)
+        return past_key_values
+
+    def forward(
+            self,
+            input_ids,
+            position_ids: Optional[torch.Tensor] = None,
+            attention_mask: Optional[torch.BoolTensor] = None,
+            full_attention_mask: Optional[torch.BoolTensor] = None,
+            past_key_values: Optional[Tuple[Tuple[torch.Tensor, torch.Tensor], ...]] = None,
+            inputs_embeds: Optional[torch.Tensor] = None,
+            use_cache: Optional[bool] = None,
+            output_hidden_states: Optional[bool] = None,
+            return_dict: Optional[bool] = None,
+    ):
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        use_cache = use_cache if use_cache is not None else self.config.use_cache
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        batch_size, seq_length = input_ids.shape
+
+        if inputs_embeds is None:
+            inputs_embeds = self.embedding(input_ids)
+
+        if self.pre_seq_len is not None:
+            if past_key_values is None:
+                past_key_values = self.get_prompt(batch_size=batch_size, device=input_ids.device,
+                                                  dtype=inputs_embeds.dtype)
+            if attention_mask is not None:
+                attention_mask = torch.cat([attention_mask.new_ones((batch_size, self.pre_seq_len)),
+                                            attention_mask], dim=-1)
+
+        if full_attention_mask is None:
+            if (attention_mask is not None and not attention_mask.all()) or (past_key_values and seq_length != 1):
+                full_attention_mask = self.get_masks(input_ids, past_key_values, padding_mask=attention_mask)
+
+        # Rotary positional embeddings
+        rotary_pos_emb = self.rotary_pos_emb(self.seq_length)
+        if position_ids is not None:
+            rotary_pos_emb = rotary_pos_emb[position_ids]
+        else:
+            rotary_pos_emb = rotary_pos_emb[None, :seq_length]
+        rotary_pos_emb = rotary_pos_emb.transpose(0, 1).contiguous()
+
+        # Run encoder.
+        hidden_states, presents, all_hidden_states, all_self_attentions = self.encoder(
+            inputs_embeds, full_attention_mask, rotary_pos_emb=rotary_pos_emb,
+            kv_caches=past_key_values, use_cache=use_cache, output_hidden_states=output_hidden_states
+        )
+
+        if not return_dict:
+            return tuple(v for v in [hidden_states, presents, all_hidden_states, all_self_attentions] if v is not None)
+
+        return BaseModelOutputWithPast(
+            last_hidden_state=hidden_states,
+            past_key_values=presents,
+            hidden_states=all_hidden_states,
+            attentions=all_self_attentions,
+        )
+
+    def quantize(self, weight_bit_width: int):
+        from .quantization import quantize
+        quantize(self.encoder, weight_bit_width)
+        return self
+
+
+class ChatGLMForConditionalGeneration(ChatGLMPreTrainedModel):
+    def __init__(self, config: ChatGLMConfig, empty_init=True, device=None):
+        super().__init__(config)
+
+        self.max_sequence_length = config.max_length
+        self.transformer = ChatGLMModel(config, empty_init=empty_init, device=device)
+        self.config = config
+        self.quantized = False
+
+        if self.config.quantization_bit:
+            self.quantize(self.config.quantization_bit, empty_init=True)
+
+    def _update_model_kwargs_for_generation(
+            self,
+            outputs: ModelOutput,
+            model_kwargs: Dict[str, Any],
+            is_encoder_decoder: bool = False,
+            standardize_cache_format: bool = False,
+    ) -> Dict[str, Any]:
+        # update past_key_values
+        model_kwargs["past_key_values"] = self._extract_past_from_model_output(
+            outputs, standardize_cache_format=standardize_cache_format
+        )
+
+        # update attention mask
+        if "attention_mask" in model_kwargs:
+            attention_mask = model_kwargs["attention_mask"]
+            model_kwargs["attention_mask"] = torch.cat(
+                [attention_mask, attention_mask.new_ones((attention_mask.shape[0], 1))], dim=-1
+            )
+
+        # update position ids
+        if "position_ids" in model_kwargs:
+            position_ids = model_kwargs["position_ids"]
+            new_position_id = position_ids[..., -1:].clone()
+            new_position_id += 1
+            model_kwargs["position_ids"] = torch.cat(
+                [position_ids, new_position_id], dim=-1
+            )
+
+        model_kwargs["is_first_forward"] = False
+        return model_kwargs
+
+    def prepare_inputs_for_generation(
+            self,
+            input_ids: torch.LongTensor,
+            past_key_values: Optional[torch.Tensor] = None,
+            attention_mask: Optional[torch.Tensor] = None,
+            position_ids: Optional[torch.Tensor] = None,
+            use_cache: Optional[bool] = None,
+            is_first_forward: bool = True,
+            **kwargs
+    ) -> dict:
+        # only last token for input_ids if past is not None
+        if position_ids is None:
+            position_ids = self.get_position_ids(input_ids, device=input_ids.device)
+        if not is_first_forward:
+            if past_key_values is not None:
+                position_ids = position_ids[..., -1:]
+                input_ids = input_ids[:, -1:]
+        return {
+            "input_ids": input_ids,
+            "past_key_values": past_key_values,
+            "position_ids": position_ids,
+            "attention_mask": attention_mask,
+            "return_last_logit": True,
+            "use_cache": use_cache
+        }
+
+    def forward(
+            self,
+            input_ids: Optional[torch.Tensor] = None,
+            position_ids: Optional[torch.Tensor] = None,
+            attention_mask: Optional[torch.Tensor] = None,
+            past_key_values: Optional[Tuple[torch.FloatTensor]] = None,
+            inputs_embeds: Optional[torch.Tensor] = None,
+            labels: Optional[torch.Tensor] = None,
+            use_cache: Optional[bool] = None,
+            output_attentions: Optional[bool] = None,
+            output_hidden_states: Optional[bool] = None,
+            return_dict: Optional[bool] = None,
+            return_last_logit: Optional[bool] = False,
+    ):
+        use_cache = use_cache if use_cache is not None else self.config.use_cache
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        transformer_outputs = self.transformer(
+            input_ids=input_ids,
+            position_ids=position_ids,
+            attention_mask=attention_mask,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        hidden_states = transformer_outputs[0]
+        if return_last_logit:
+            hidden_states = hidden_states[-1:]
+        lm_logits = self.transformer.output_layer(hidden_states)
+        lm_logits = lm_logits.transpose(0, 1).contiguous()
+
+        loss = None
+        if labels is not None:
+            lm_logits = lm_logits.to(torch.float32)
+
+            # Shift so that tokens < n predict n
+            shift_logits = lm_logits[..., :-1, :].contiguous()
+            shift_labels = labels[..., 1:].contiguous()
+            # Flatten the tokens
+            loss_fct = CrossEntropyLoss(ignore_index=-100)
+            loss = loss_fct(shift_logits.view(-1, shift_logits.size(-1)), shift_labels.view(-1))
+
+            lm_logits = lm_logits.to(hidden_states.dtype)
+            loss = loss.to(hidden_states.dtype)
+
+        if not return_dict:
+            output = (lm_logits,) + transformer_outputs[1:]
+            return ((loss,) + output) if loss is not None else output
+
+        return CausalLMOutputWithPast(
+            loss=loss,
+            logits=lm_logits,
+            past_key_values=transformer_outputs.past_key_values,
+            hidden_states=transformer_outputs.hidden_states,
+            attentions=transformer_outputs.attentions,
+        )
+
+    @staticmethod
+    def _reorder_cache(
+            past: Tuple[Tuple[torch.Tensor, torch.Tensor], ...], beam_idx: torch.LongTensor
+    ) -> Tuple[Tuple[torch.Tensor, torch.Tensor], ...]:
+        """
+        This function is used to re-order the `past_key_values` cache if [`~PreTrainedModel.beam_search`] or
+        [`~PreTrainedModel.beam_sample`] is called. This is required to match `past_key_values` with the correct
+        beam_idx at every generation step.
+
+        Output shares the same memory storage as `past`.
+        """
+        return tuple(
+            (
+                layer_past[0].index_select(1, beam_idx.to(layer_past[0].device)),
+                layer_past[1].index_select(1, beam_idx.to(layer_past[1].device)),
+            )
+            for layer_past in past
+        )
+
+    def process_response(self, output, history):
+        content = ""
+        history = deepcopy(history)
+        for response in output.split("<|assistant|>"):
+            metadata, content = response.split("\n", maxsplit=1)
+            if not metadata.strip():
+                content = content.strip()
+                history.append({"role": "assistant", "metadata": metadata, "content": content})
+                content = content.replace("[[训练时间]]", "2023年")
+            else:
+                history.append({"role": "assistant", "metadata": metadata, "content": content})
+                if history[0]["role"] == "system" and "tools" in history[0]:
+                    content = "\n".join(content.split("\n")[1:-1])
+                    def tool_call(**kwargs):
+                        return kwargs
+                    parameters = eval(content)
+                    content = {"name": metadata.strip(), "parameters": parameters}
+                else:
+                    content = {"name": metadata.strip(), "content": content}
+        return content, history
+
+    @torch.inference_mode()
+    def chat(self, tokenizer, query: str, history: List[Dict] = None, role: str = "user",
+             max_length: int = 8192, num_beams=1, do_sample=True, top_p=0.8, temperature=0.8, logits_processor=None,
+             **kwargs):
+        if history is None:
+            history = []
+        if logits_processor is None:
+            logits_processor = LogitsProcessorList()
+        logits_processor.append(InvalidScoreLogitsProcessor())
+        gen_kwargs = {"max_length": max_length, "num_beams": num_beams, "do_sample": do_sample, "top_p": top_p,
+                      "temperature": temperature, "logits_processor": logits_processor, **kwargs}
+        inputs = tokenizer.build_chat_input(query, history=history, role=role)
+        inputs = inputs.to(self.device)
+        eos_token_id = [tokenizer.eos_token_id, tokenizer.get_command("<|user|>"),
+                        tokenizer.get_command("<|observation|>")]
+        outputs = self.generate(**inputs, **gen_kwargs, eos_token_id=eos_token_id)
+        outputs = outputs.tolist()[0][len(inputs["input_ids"][0]):-1]
+        response = tokenizer.decode(outputs)
+        history.append({"role": role, "content": query})
+        response, history = self.process_response(response, history)
+        return response, history
+
+    @torch.inference_mode()
+    def stream_chat(self, tokenizer, query: str, history: List[Dict] = None, role: str = "user",
+                    past_key_values=None,max_length: int = 8192, do_sample=True, top_p=0.8, temperature=0.8,
+                    logits_processor=None, return_past_key_values=False, **kwargs):
+        if history is None:
+            history = []
+        if logits_processor is None:
+            logits_processor = LogitsProcessorList()
+        logits_processor.append(InvalidScoreLogitsProcessor())
+        eos_token_id = [tokenizer.eos_token_id, tokenizer.get_command("<|user|>"),
+                        tokenizer.get_command("<|observation|>")]
+        gen_kwargs = {"max_length": max_length, "do_sample": do_sample, "top_p": top_p,
+                      "temperature": temperature, "logits_processor": logits_processor, **kwargs}
+        if past_key_values is None:
+            inputs = tokenizer.build_chat_input(query, history=history, role=role)
+        else:
+            inputs = tokenizer.build_chat_input(query, role=role)
+        inputs = inputs.to(self.device)
+        if past_key_values is not None:
+            past_length = past_key_values[0][0].shape[0]
+            if self.transformer.pre_seq_len is not None:
+                past_length -= self.transformer.pre_seq_len
+            inputs.position_ids += past_length
+            attention_mask = inputs.attention_mask
+            attention_mask = torch.cat((attention_mask.new_ones(1, past_length), attention_mask), dim=1)
+            inputs['attention_mask'] = attention_mask
+        history.append({"role": role, "content": query})
+        for outputs in self.stream_generate(**inputs, past_key_values=past_key_values,
+                                            eos_token_id=eos_token_id, return_past_key_values=return_past_key_values,
+                                            **gen_kwargs):
+            if return_past_key_values:
+                outputs, past_key_values = outputs
+            outputs = outputs.tolist()[0][len(inputs["input_ids"][0]):-1]
+            response = tokenizer.decode(outputs)
+            if response and response[-1] != "�":
+                response, new_history = self.process_response(response, history)
+                if return_past_key_values:
+                    yield response, new_history, past_key_values
+                else:
+                    yield response, new_history
+
+    @torch.inference_mode()
+    def stream_generate(
+            self,
+            input_ids,
+            generation_config: Optional[GenerationConfig] = None,
+            logits_processor: Optional[LogitsProcessorList] = None,
+            stopping_criteria: Optional[StoppingCriteriaList] = None,
+            prefix_allowed_tokens_fn: Optional[Callable[[int, torch.Tensor], List[int]]] = None,
+            return_past_key_values=False,
+            **kwargs,
+    ):
+        batch_size, input_ids_seq_length = input_ids.shape[0], input_ids.shape[-1]
+
+        if generation_config is None:
+            generation_config = self.generation_config
+        generation_config = copy.deepcopy(generation_config)
+        model_kwargs = generation_config.update(**kwargs)
+        model_kwargs["use_cache"] = generation_config.use_cache
+        bos_token_id, eos_token_id = generation_config.bos_token_id, generation_config.eos_token_id
+
+        if isinstance(eos_token_id, int):
+            eos_token_id = [eos_token_id]
+        eos_token_id_tensor = torch.tensor(eos_token_id).to(input_ids.device) if eos_token_id is not None else None
+
+        has_default_max_length = kwargs.get("max_length") is None and generation_config.max_length is not None
+        if has_default_max_length and generation_config.max_new_tokens is None:
+            warnings.warn(
+                f"Using `max_length`'s default ({generation_config.max_length}) to control the generation length. "
+                "This behaviour is deprecated and will be removed from the config in v5 of Transformers -- we"
+                " recommend using `max_new_tokens` to control the maximum length of the generation.",
+                UserWarning,
+            )
+        elif generation_config.max_new_tokens is not None:
+            generation_config.max_length = generation_config.max_new_tokens + input_ids_seq_length
+            if not has_default_max_length:
+                logger.warn(
+                    f"Both `max_new_tokens` (={generation_config.max_new_tokens}) and `max_length`(="
+                    f"{generation_config.max_length}) seem to have been set. `max_new_tokens` will take precedence. "
+                    "Please refer to the documentation for more information. "
+                    "(https://huggingface.co/docs/transformers/main/en/main_classes/text_generation)",
+                    UserWarning,
+                )
+
+        if input_ids_seq_length >= generation_config.max_length:
+            input_ids_string = "decoder_input_ids" if self.config.is_encoder_decoder else "input_ids"
+            logger.warning(
+                f"Input length of {input_ids_string} is {input_ids_seq_length}, but `max_length` is set to"
+                f" {generation_config.max_length}. This can lead to unexpected behavior. You should consider"
+                " increasing `max_new_tokens`."
+            )
+
+        # 2. Set generation parameters if not already defined
+        logits_processor = logits_processor if logits_processor is not None else LogitsProcessorList()
+        stopping_criteria = stopping_criteria if stopping_criteria is not None else StoppingCriteriaList()
+
+        logits_processor = self._get_logits_processor(
+            generation_config=generation_config,
+            input_ids_seq_length=input_ids_seq_length,
+            encoder_input_ids=input_ids,
+            prefix_allowed_tokens_fn=prefix_allowed_tokens_fn,
+            logits_processor=logits_processor,
+        )
+
+        stopping_criteria = self._get_stopping_criteria(
+            generation_config=generation_config, stopping_criteria=stopping_criteria
+        )
+        logits_warper = self._get_logits_warper(generation_config)
+
+        unfinished_sequences = input_ids.new(input_ids.shape[0]).fill_(1)
+        scores = None
+        while True:
+            model_inputs = self.prepare_inputs_for_generation(input_ids, **model_kwargs)
+            # forward pass to get next token
+            outputs = self(
+                **model_inputs,
+                return_dict=True,
+                output_attentions=False,
+                output_hidden_states=False,
+            )
+
+            next_token_logits = outputs.logits[:, -1, :]
+
+            # pre-process distribution
+            next_token_scores = logits_processor(input_ids, next_token_logits)
+            next_token_scores = logits_warper(input_ids, next_token_scores)
+
+            # sample
+            probs = nn.functional.softmax(next_token_scores, dim=-1)
+            if generation_config.do_sample:
+                next_tokens = torch.multinomial(probs, num_samples=1).squeeze(1)
+            else:
+                next_tokens = torch.argmax(probs, dim=-1)
+            # update generated ids, model inputs, and length for next step
+            input_ids = torch.cat([input_ids, next_tokens[:, None]], dim=-1)
+            model_kwargs = self._update_model_kwargs_for_generation(
+                outputs, model_kwargs, is_encoder_decoder=self.config.is_encoder_decoder
+            )
+            unfinished_sequences = unfinished_sequences.mul(
+                next_tokens.tile(eos_token_id_tensor.shape[0], 1).ne(eos_token_id_tensor.unsqueeze(1)).prod(dim=0)
+            )
+            if return_past_key_values:
+                yield input_ids, outputs.past_key_values
+            else:
+                yield input_ids
+            # stop when each sentence is finished, or if we exceed the maximum length
+            if unfinished_sequences.max() == 0 or stopping_criteria(input_ids, scores):
+                break
+
+    def quantize(self, bits: int, empty_init=False, device=None, **kwargs):
+        if bits == 0:
+            return
+
+        from .quantization import quantize
+
+        if self.quantized:
+            logger.info("Already quantized.")
+            return self
+
+        self.quantized = True
+
+        self.config.quantization_bit = bits
+
+        self.transformer.encoder = quantize(self.transformer.encoder, bits, empty_init=empty_init, device=device,
+                                            **kwargs)
+        return self
+
+
+class ChatGLMForSequenceClassification(ChatGLMPreTrainedModel):
+    def __init__(self, config: ChatGLMConfig, empty_init=True, device=None):
+        super().__init__(config)
+
+        self.num_labels = config.num_labels
+        self.transformer = ChatGLMModel(config, empty_init=empty_init, device=device)
+
+        self.classifier_head = nn.Linear(config.hidden_size, config.num_labels, bias=True, dtype=torch.half)
+        if config.classifier_dropout is not None:
+            self.dropout = nn.Dropout(config.classifier_dropout)
+        else:
+            self.dropout = None
+        self.config = config
+
+        if self.config.quantization_bit:
+            self.quantize(self.config.quantization_bit, empty_init=True)
+
+    def forward(
+            self,
+            input_ids: Optional[torch.LongTensor] = None,
+            position_ids: Optional[torch.LongTensor] = None,
+            attention_mask: Optional[torch.Tensor] = None,
+            full_attention_mask: Optional[torch.Tensor] = None,
+            past_key_values: Optional[Tuple[Tuple[torch.Tensor, torch.Tensor], ...]] = None,
+            inputs_embeds: Optional[torch.LongTensor] = None,
+            labels: Optional[torch.LongTensor] = None,
+            use_cache: Optional[bool] = None,
+            output_hidden_states: Optional[bool] = None,
+            return_dict: Optional[bool] = None,
+    ) -> Union[Tuple[torch.Tensor, ...], SequenceClassifierOutputWithPast]:
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        transformer_outputs = self.transformer(
+            input_ids=input_ids,
+            position_ids=position_ids,
+            attention_mask=attention_mask,
+            full_attention_mask=full_attention_mask,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        hidden_states = transformer_outputs[0]
+        pooled_hidden_states = hidden_states[-1]
+        if self.dropout is not None:
+            pooled_hidden_states = self.dropout(pooled_hidden_states)
+        logits = self.classifier_head(pooled_hidden_states)
+
+        loss = None
+        if labels is not None:
+            if self.config.problem_type is None:
+                if self.num_labels == 1:
+                    self.config.problem_type = "regression"
+                elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
+                    self.config.problem_type = "single_label_classification"
+                else:
+                    self.config.problem_type = "multi_label_classification"
+
+            if self.config.problem_type == "regression":
+                loss_fct = MSELoss()
+                if self.num_labels == 1:
+                    loss = loss_fct(logits.squeeze().float(), labels.squeeze())
+                else:
+                    loss = loss_fct(logits.float(), labels)
+            elif self.config.problem_type == "single_label_classification":
+                loss_fct = CrossEntropyLoss()
+                loss = loss_fct(logits.view(-1, self.num_labels).float(), labels.view(-1))
+            elif self.config.problem_type == "multi_label_classification":
+                loss_fct = BCEWithLogitsLoss()
+                loss = loss_fct(logits.float(), labels.view(-1, self.num_labels))
+
+        if not return_dict:
+            output = (logits,) + transformer_outputs[1:]
+            return ((loss,) + output) if loss is not None else output
+
+        return SequenceClassifierOutputWithPast(
+            loss=loss,
+            logits=logits,
+            past_key_values=transformer_outputs.past_key_values,
+            hidden_states=transformer_outputs.hidden_states,
+            attentions=transformer_outputs.attentions,
+        )
diff --git a/chatglm3-6b/pytorch_model-00001-of-00007.bin b/chatglm3-6b/pytorch_model-00001-of-00007.bin
new file mode 100644
index 0000000000000000000000000000000000000000..60684369c9662f8bec19e84335f5b223df729829
--- /dev/null
+++ b/chatglm3-6b/pytorch_model-00001-of-00007.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d5567466e89625dbd10e51c69a02982f233a10108cf232a379defdbb065ae0b
+size 1827781090
diff --git a/chatglm3-6b/pytorch_model-00002-of-00007.bin b/chatglm3-6b/pytorch_model-00002-of-00007.bin
new file mode 100644
index 0000000000000000000000000000000000000000..757ecc50e39f03d6aae2fee4da59b1f109685b10
--- /dev/null
+++ b/chatglm3-6b/pytorch_model-00002-of-00007.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ad41534016ac4a2431fa2d4b08efbe28f963744135ec52b2ea13cc86730fa2a
+size 1968299480
diff --git a/chatglm3-6b/pytorch_model-00003-of-00007.bin b/chatglm3-6b/pytorch_model-00003-of-00007.bin
new file mode 100644
index 0000000000000000000000000000000000000000..1100399a10901ffefc26b928c3cdbbcd57c0649d
--- /dev/null
+++ b/chatglm3-6b/pytorch_model-00003-of-00007.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2be9b17c332a8345e787953d4879caee4747ad4b263f013aa3c4654c94c3d24
+size 1927415036
diff --git a/chatglm3-6b/pytorch_model-00004-of-00007.bin b/chatglm3-6b/pytorch_model-00004-of-00007.bin
new file mode 100644
index 0000000000000000000000000000000000000000..bf75214fd965edb4e9a9b787bd0f5660002992d0
--- /dev/null
+++ b/chatglm3-6b/pytorch_model-00004-of-00007.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5526891e1b4c8edd2e3688df5156aa107e2140fe7e3f9d0d54f9cbe3b6ee3b5
+size 1815225998
diff --git a/chatglm3-6b/pytorch_model-00005-of-00007.bin b/chatglm3-6b/pytorch_model-00005-of-00007.bin
new file mode 100644
index 0000000000000000000000000000000000000000..8dd945aa4e92570245a6997ec799c64dba5892b2
--- /dev/null
+++ b/chatglm3-6b/pytorch_model-00005-of-00007.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84bb18a476f74beaf4d941733bd1c475791eba799b228f78d0165de989cb7a40
+size 1968299544
diff --git a/chatglm3-6b/pytorch_model-00006-of-00007.bin b/chatglm3-6b/pytorch_model-00006-of-00007.bin
new file mode 100644
index 0000000000000000000000000000000000000000..ff0ff392ba4b2d2b77cf4e282efdcfcf4eca5d4f
--- /dev/null
+++ b/chatglm3-6b/pytorch_model-00006-of-00007.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1181875a2dc30fba968d72d0fc4628b9a60d3866bf680eb14b9822b5b504830f
+size 1927415036
diff --git a/chatglm3-6b/pytorch_model-00007-of-00007.bin b/chatglm3-6b/pytorch_model-00007-of-00007.bin
new file mode 100644
index 0000000000000000000000000000000000000000..cc2f19e83aebaf89a623891a11ceb1e72f87985a
--- /dev/null
+++ b/chatglm3-6b/pytorch_model-00007-of-00007.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1271b638dc0a88206d1c7a51bcaa862410eb56f3e59fd0068a96e96cb5e3f4f0
+size 1052808542
diff --git a/chatglm3-6b/pytorch_model.bin.index.json b/chatglm3-6b/pytorch_model.bin.index.json
new file mode 100644
index 0000000000000000000000000000000000000000..cd6a94a411934e8cc04e271f169a68754d14733d
--- /dev/null
+++ b/chatglm3-6b/pytorch_model.bin.index.json
@@ -0,0 +1,207 @@
+{
+  "metadata": {
+    "total_size": 12487168064
+  },
+  "weight_map": {
+    "transformer.embedding.word_embeddings.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.final_layernorm.weight": "pytorch_model-00007-of-00007.bin",
+    "transformer.encoder.layers.0.input_layernorm.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.0.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.0.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.0.post_attention_layernorm.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.0.self_attention.dense.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.0.self_attention.query_key_value.bias": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.0.self_attention.query_key_value.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.1.input_layernorm.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.1.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.1.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.1.post_attention_layernorm.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.1.self_attention.dense.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.1.self_attention.query_key_value.bias": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.1.self_attention.query_key_value.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.10.input_layernorm.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.10.mlp.dense_4h_to_h.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.10.mlp.dense_h_to_4h.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.10.post_attention_layernorm.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.10.self_attention.dense.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.10.self_attention.query_key_value.bias": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.10.self_attention.query_key_value.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.11.input_layernorm.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.11.mlp.dense_4h_to_h.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.11.mlp.dense_h_to_4h.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.11.post_attention_layernorm.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.11.self_attention.dense.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.11.self_attention.query_key_value.bias": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.11.self_attention.query_key_value.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.12.input_layernorm.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.12.mlp.dense_4h_to_h.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.12.mlp.dense_h_to_4h.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.12.post_attention_layernorm.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.12.self_attention.dense.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.12.self_attention.query_key_value.bias": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.12.self_attention.query_key_value.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.13.input_layernorm.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.13.mlp.dense_4h_to_h.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.13.mlp.dense_h_to_4h.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.13.post_attention_layernorm.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.13.self_attention.dense.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.13.self_attention.query_key_value.bias": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.13.self_attention.query_key_value.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.14.input_layernorm.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.14.mlp.dense_4h_to_h.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.14.mlp.dense_h_to_4h.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.14.post_attention_layernorm.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.14.self_attention.dense.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.14.self_attention.query_key_value.bias": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.14.self_attention.query_key_value.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.15.input_layernorm.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.15.mlp.dense_4h_to_h.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.15.mlp.dense_h_to_4h.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.15.post_attention_layernorm.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.15.self_attention.dense.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.15.self_attention.query_key_value.bias": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.15.self_attention.query_key_value.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.16.input_layernorm.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.16.mlp.dense_4h_to_h.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.16.mlp.dense_h_to_4h.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.16.post_attention_layernorm.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.16.self_attention.dense.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.16.self_attention.query_key_value.bias": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.16.self_attention.query_key_value.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.17.input_layernorm.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.17.mlp.dense_4h_to_h.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.17.mlp.dense_h_to_4h.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.17.post_attention_layernorm.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.17.self_attention.dense.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.17.self_attention.query_key_value.bias": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.17.self_attention.query_key_value.weight": "pytorch_model-00004-of-00007.bin",
+    "transformer.encoder.layers.18.input_layernorm.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.18.mlp.dense_4h_to_h.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.18.mlp.dense_h_to_4h.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.18.post_attention_layernorm.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.18.self_attention.dense.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.18.self_attention.query_key_value.bias": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.18.self_attention.query_key_value.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.19.input_layernorm.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.19.mlp.dense_4h_to_h.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.19.mlp.dense_h_to_4h.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.19.post_attention_layernorm.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.19.self_attention.dense.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.19.self_attention.query_key_value.bias": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.19.self_attention.query_key_value.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.2.input_layernorm.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.2.mlp.dense_4h_to_h.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.2.mlp.dense_h_to_4h.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.2.post_attention_layernorm.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.2.self_attention.dense.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.2.self_attention.query_key_value.bias": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.2.self_attention.query_key_value.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.20.input_layernorm.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.20.mlp.dense_4h_to_h.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.20.mlp.dense_h_to_4h.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.20.post_attention_layernorm.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.20.self_attention.dense.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.20.self_attention.query_key_value.bias": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.20.self_attention.query_key_value.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.21.input_layernorm.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.21.mlp.dense_4h_to_h.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.21.mlp.dense_h_to_4h.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.21.post_attention_layernorm.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.21.self_attention.dense.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.21.self_attention.query_key_value.bias": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.21.self_attention.query_key_value.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.22.input_layernorm.weight": "pytorch_model-00005-of-00007.bin",
+    "transformer.encoder.layers.22.mlp.dense_4h_to_h.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.22.mlp.dense_h_to_4h.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.22.post_attention_layernorm.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.22.self_attention.dense.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.22.self_attention.query_key_value.bias": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.22.self_attention.query_key_value.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.23.input_layernorm.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.23.mlp.dense_4h_to_h.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.23.mlp.dense_h_to_4h.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.23.post_attention_layernorm.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.23.self_attention.dense.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.23.self_attention.query_key_value.bias": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.23.self_attention.query_key_value.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.24.input_layernorm.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.24.mlp.dense_4h_to_h.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.24.mlp.dense_h_to_4h.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.24.post_attention_layernorm.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.24.self_attention.dense.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.24.self_attention.query_key_value.bias": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.24.self_attention.query_key_value.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.25.input_layernorm.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.25.mlp.dense_4h_to_h.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.25.mlp.dense_h_to_4h.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.25.post_attention_layernorm.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.25.self_attention.dense.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.25.self_attention.query_key_value.bias": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.25.self_attention.query_key_value.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.26.input_layernorm.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.26.mlp.dense_4h_to_h.weight": "pytorch_model-00007-of-00007.bin",
+    "transformer.encoder.layers.26.mlp.dense_h_to_4h.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.26.post_attention_layernorm.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.26.self_attention.dense.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.26.self_attention.query_key_value.bias": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.26.self_attention.query_key_value.weight": "pytorch_model-00006-of-00007.bin",
+    "transformer.encoder.layers.27.input_layernorm.weight": "pytorch_model-00007-of-00007.bin",
+    "transformer.encoder.layers.27.mlp.dense_4h_to_h.weight": "pytorch_model-00007-of-00007.bin",
+    "transformer.encoder.layers.27.mlp.dense_h_to_4h.weight": "pytorch_model-00007-of-00007.bin",
+    "transformer.encoder.layers.27.post_attention_layernorm.weight": "pytorch_model-00007-of-00007.bin",
+    "transformer.encoder.layers.27.self_attention.dense.weight": "pytorch_model-00007-of-00007.bin",
+    "transformer.encoder.layers.27.self_attention.query_key_value.bias": "pytorch_model-00007-of-00007.bin",
+    "transformer.encoder.layers.27.self_attention.query_key_value.weight": "pytorch_model-00007-of-00007.bin",
+    "transformer.encoder.layers.3.input_layernorm.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.3.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.3.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.3.post_attention_layernorm.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.3.self_attention.dense.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.3.self_attention.query_key_value.bias": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.3.self_attention.query_key_value.weight": "pytorch_model-00001-of-00007.bin",
+    "transformer.encoder.layers.4.input_layernorm.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.4.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.4.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.4.post_attention_layernorm.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.4.self_attention.dense.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.4.self_attention.query_key_value.bias": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.4.self_attention.query_key_value.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.5.input_layernorm.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.5.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.5.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.5.post_attention_layernorm.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.5.self_attention.dense.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.5.self_attention.query_key_value.bias": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.5.self_attention.query_key_value.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.6.input_layernorm.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.6.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.6.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.6.post_attention_layernorm.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.6.self_attention.dense.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.6.self_attention.query_key_value.bias": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.6.self_attention.query_key_value.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.7.input_layernorm.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.7.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.7.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.7.post_attention_layernorm.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.7.self_attention.dense.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.7.self_attention.query_key_value.bias": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.7.self_attention.query_key_value.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.8.input_layernorm.weight": "pytorch_model-00002-of-00007.bin",
+    "transformer.encoder.layers.8.mlp.dense_4h_to_h.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.8.mlp.dense_h_to_4h.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.8.post_attention_layernorm.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.8.self_attention.dense.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.8.self_attention.query_key_value.bias": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.8.self_attention.query_key_value.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.9.input_layernorm.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.9.mlp.dense_4h_to_h.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.9.mlp.dense_h_to_4h.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.9.post_attention_layernorm.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.9.self_attention.dense.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.9.self_attention.query_key_value.bias": "pytorch_model-00003-of-00007.bin",
+    "transformer.encoder.layers.9.self_attention.query_key_value.weight": "pytorch_model-00003-of-00007.bin",
+    "transformer.output_layer.weight": "pytorch_model-00007-of-00007.bin",
+    "transformer.rotary_pos_emb.inv_freq": "pytorch_model-00001-of-00007.bin"
+  }
+}
\ No newline at end of file
diff --git a/chatglm3-6b/quantization.py b/chatglm3-6b/quantization.py
new file mode 100644
index 0000000000000000000000000000000000000000..cb95bfe82b203ff6a2aa962326d2c7a438d6a52f
--- /dev/null
+++ b/chatglm3-6b/quantization.py
@@ -0,0 +1,188 @@
+from torch.nn import Linear
+from torch.nn.parameter import Parameter
+
+import bz2
+import torch
+import base64
+import ctypes
+from transformers.utils import logging
+
+from typing import List
+from functools import partial
+
+logger = logging.get_logger(__name__)
+
+try:
+    from cpm_kernels.kernels.base import LazyKernelCModule, KernelFunction, round_up
+
+    class Kernel:
+        def __init__(self, code: bytes, function_names: List[str]):
+            self.code = code
+            self._function_names = function_names
+            self._cmodule = LazyKernelCModule(self.code)
+
+            for name in self._function_names:
+                setattr(self, name, KernelFunction(self._cmodule, name))
+
+    quantization_code = "$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"
+
+    kernels = Kernel(
+        bz2.decompress(base64.b64decode(quantization_code)),
+        [
+            "int4WeightCompression",
+            "int4WeightExtractionFloat",
+            "int4WeightExtractionHalf",
+            "int8WeightExtractionFloat",
+            "int8WeightExtractionHalf",
+        ],
+    )
+except Exception as exception:
+    kernels = None
+    logger.warning("Failed to load cpm_kernels:" + str(exception))
+
+
+class W8A16Linear(torch.autograd.Function):
+    @staticmethod
+    def forward(ctx, inp: torch.Tensor, quant_w: torch.Tensor, scale_w: torch.Tensor, weight_bit_width):
+        ctx.inp_shape = inp.size()
+        ctx.weight_bit_width = weight_bit_width
+        out_features = quant_w.size(0)
+        inp = inp.contiguous().view(-1, inp.size(-1))
+        weight = extract_weight_to_half(quant_w, scale_w, weight_bit_width)
+        ctx.weight_shape = weight.size()
+        output = inp.mm(weight.t())
+        ctx.save_for_backward(inp, quant_w, scale_w)
+        return output.view(*(ctx.inp_shape[:-1] + (out_features,)))
+
+    @staticmethod
+    def backward(ctx, grad_output: torch.Tensor):
+        inp, quant_w, scale_w = ctx.saved_tensors
+        weight = extract_weight_to_half(quant_w, scale_w, ctx.weight_bit_width)
+        grad_output = grad_output.contiguous().view(-1, weight.size(0))
+        grad_input = grad_output.mm(weight)
+        grad_weight = grad_output.t().mm(inp)
+        return grad_input.view(ctx.inp_shape), grad_weight.view(ctx.weight_shape), None, None
+
+
+def compress_int4_weight(weight: torch.Tensor):  # (n, m)
+    with torch.cuda.device(weight.device):
+        n, m = weight.size(0), weight.size(1)
+        assert m % 2 == 0
+        m = m // 2
+        out = torch.empty(n, m, dtype=torch.int8, device="cuda")
+        stream = torch.cuda.current_stream()
+
+        gridDim = (n, 1, 1)
+        blockDim = (min(round_up(m, 32), 1024), 1, 1)
+
+        kernels.int4WeightCompression(
+            gridDim,
+            blockDim,
+            0,
+            stream,
+            [ctypes.c_void_p(weight.data_ptr()), ctypes.c_void_p(out.data_ptr()), ctypes.c_int32(n), ctypes.c_int32(m)],
+        )
+        return out
+
+
+def extract_weight_to_half(weight: torch.Tensor, scale_list: torch.Tensor, source_bit_width: int):
+    assert scale_list.dtype in [torch.half, torch.bfloat16]
+    assert weight.dtype in [torch.int8]
+    if source_bit_width == 8:
+        return weight.to(scale_list.dtype) * scale_list[:, None]
+    elif source_bit_width == 4:
+        func = (
+            kernels.int4WeightExtractionHalf if scale_list.dtype == torch.half else kernels.int4WeightExtractionBFloat16
+        )
+    else:
+        assert False, "Unsupported bit-width"
+
+    with torch.cuda.device(weight.device):
+        n, m = weight.size(0), weight.size(1)
+        out = torch.empty(n, m * (8 // source_bit_width), dtype=scale_list.dtype, device="cuda")
+        stream = torch.cuda.current_stream()
+
+        gridDim = (n, 1, 1)
+        blockDim = (min(round_up(m, 32), 1024), 1, 1)
+
+        func(
+            gridDim,
+            blockDim,
+            0,
+            stream,
+            [
+                ctypes.c_void_p(weight.data_ptr()),
+                ctypes.c_void_p(scale_list.data_ptr()),
+                ctypes.c_void_p(out.data_ptr()),
+                ctypes.c_int32(n),
+                ctypes.c_int32(m),
+            ],
+        )
+        return out
+
+
+class QuantizedLinear(torch.nn.Module):
+    def __init__(self, weight_bit_width: int, weight, bias=None, device="cpu", dtype=None, empty_init=False, *args,
+                 **kwargs):
+        super().__init__()
+        self.weight_bit_width = weight_bit_width
+
+        shape = weight.shape
+
+        if weight is None or empty_init:
+            self.weight = torch.empty(shape[0], shape[1] * weight_bit_width // 8, dtype=torch.int8, device=device)
+            self.weight_scale = torch.empty(shape[0], dtype=dtype, device=device)
+        else:
+            self.weight_scale = weight.abs().max(dim=-1).values / ((2 ** (weight_bit_width - 1)) - 1)
+            self.weight = torch.round(weight / self.weight_scale[:, None]).to(torch.int8)
+            if weight_bit_width == 4:
+                self.weight = compress_int4_weight(self.weight)
+
+        self.weight = Parameter(self.weight.to(device), requires_grad=False)
+        self.weight_scale = Parameter(self.weight_scale.to(device), requires_grad=False)
+        self.bias = Parameter(bias.to(device), requires_grad=False) if bias is not None else None
+
+    def forward(self, input):
+        output = W8A16Linear.apply(input, self.weight, self.weight_scale, self.weight_bit_width)
+        if self.bias is not None:
+            output = output + self.bias
+        return output
+
+
+def quantize(model, weight_bit_width, empty_init=False, device=None):
+    """Replace fp16 linear with quantized linear"""
+    for layer in model.layers:
+        layer.self_attention.query_key_value = QuantizedLinear(
+            weight_bit_width=weight_bit_width,
+            weight=layer.self_attention.query_key_value.weight.to(torch.cuda.current_device()),
+            bias=layer.self_attention.query_key_value.bias,
+            dtype=layer.self_attention.query_key_value.weight.dtype,
+            device=layer.self_attention.query_key_value.weight.device if device is None else device,
+            empty_init=empty_init
+        )
+        layer.self_attention.dense = QuantizedLinear(
+            weight_bit_width=weight_bit_width,
+            weight=layer.self_attention.dense.weight.to(torch.cuda.current_device()),
+            bias=layer.self_attention.dense.bias,
+            dtype=layer.self_attention.dense.weight.dtype,
+            device=layer.self_attention.dense.weight.device if device is None else device,
+            empty_init=empty_init
+        )
+        layer.mlp.dense_h_to_4h = QuantizedLinear(
+            weight_bit_width=weight_bit_width,
+            weight=layer.mlp.dense_h_to_4h.weight.to(torch.cuda.current_device()),
+            bias=layer.mlp.dense_h_to_4h.bias,
+            dtype=layer.mlp.dense_h_to_4h.weight.dtype,
+            device=layer.mlp.dense_h_to_4h.weight.device if device is None else device,
+            empty_init=empty_init
+        )
+        layer.mlp.dense_4h_to_h = QuantizedLinear(
+            weight_bit_width=weight_bit_width,
+            weight=layer.mlp.dense_4h_to_h.weight.to(torch.cuda.current_device()),
+            bias=layer.mlp.dense_4h_to_h.bias,
+            dtype=layer.mlp.dense_4h_to_h.weight.dtype,
+            device=layer.mlp.dense_4h_to_h.weight.device if device is None else device,
+            empty_init=empty_init
+        )
+
+    return model
diff --git a/chatglm3-6b/tokenization_chatglm.py b/chatglm3-6b/tokenization_chatglm.py
new file mode 100644
index 0000000000000000000000000000000000000000..50e44b05e4b3e54d2f1c3f0cab8247ea53a7d4e5
--- /dev/null
+++ b/chatglm3-6b/tokenization_chatglm.py
@@ -0,0 +1,300 @@
+import json
+import os
+import re
+from typing import List, Optional, Union, Dict
+from sentencepiece import SentencePieceProcessor
+from transformers import PreTrainedTokenizer
+from transformers.utils import logging, PaddingStrategy
+from transformers.tokenization_utils_base import EncodedInput, BatchEncoding
+
+
+class SPTokenizer:
+    def __init__(self, model_path: str):
+        # reload tokenizer
+        assert os.path.isfile(model_path), model_path
+        self.sp_model = SentencePieceProcessor(model_file=model_path)
+
+        # BOS / EOS token IDs
+        self.n_words: int = self.sp_model.vocab_size()
+        self.bos_id: int = self.sp_model.bos_id()
+        self.eos_id: int = self.sp_model.eos_id()
+        self.pad_id: int = self.sp_model.unk_id()
+        assert self.sp_model.vocab_size() == self.sp_model.get_piece_size()
+
+        role_special_tokens = ["<|system|>", "<|user|>", "<|assistant|>", "<|observation|>"]
+        special_tokens = ["[MASK]", "[gMASK]", "[sMASK]", "sop", "eop"] + role_special_tokens
+        self.special_tokens = {}
+        self.index_special_tokens = {}
+        for token in special_tokens:
+            self.special_tokens[token] = self.n_words
+            self.index_special_tokens[self.n_words] = token
+            self.n_words += 1
+        self.role_special_token_expression = "|".join([re.escape(token) for token in role_special_tokens])
+
+    def tokenize(self, s: str, encode_special_tokens=False):
+        if encode_special_tokens:
+            last_index = 0
+            t = []
+            for match in re.finditer(self.role_special_token_expression, s):
+                if last_index < match.start():
+                    t.extend(self.sp_model.EncodeAsPieces(s[last_index:match.start()]))
+                t.append(s[match.start():match.end()])
+                last_index = match.end()
+            if last_index < len(s):
+                t.extend(self.sp_model.EncodeAsPieces(s[last_index:]))
+            return t
+        else:
+            return self.sp_model.EncodeAsPieces(s)
+
+    def encode(self, s: str, bos: bool = False, eos: bool = False) -> List[int]:
+        assert type(s) is str
+        t = self.sp_model.encode(s)
+        if bos:
+            t = [self.bos_id] + t
+        if eos:
+            t = t + [self.eos_id]
+        return t
+
+    def decode(self, t: List[int]) -> str:
+        text, buffer = "", []
+        for token in t:
+            if token in self.index_special_tokens:
+                if buffer:
+                    text += self.sp_model.decode(buffer)
+                    buffer = []
+                text += self.index_special_tokens[token]
+            else:
+                buffer.append(token)
+        if buffer:
+            text += self.sp_model.decode(buffer)
+        return text
+
+    def decode_tokens(self, tokens: List[str]) -> str:
+        text = self.sp_model.DecodePieces(tokens)
+        return text
+
+    def convert_token_to_id(self, token):
+        """ Converts a token (str) in an id using the vocab. """
+        if token in self.special_tokens:
+            return self.special_tokens[token]
+        return self.sp_model.PieceToId(token)
+
+    def convert_id_to_token(self, index):
+        """Converts an index (integer) in a token (str) using the vocab."""
+        if index in self.index_special_tokens:
+            return self.index_special_tokens[index]
+        if index in [self.eos_id, self.bos_id, self.pad_id] or index < 0:
+            return ""
+        return self.sp_model.IdToPiece(index)
+
+
+class ChatGLMTokenizer(PreTrainedTokenizer):
+    vocab_files_names = {"vocab_file": "tokenizer.model"}
+
+    model_input_names = ["input_ids", "attention_mask", "position_ids"]
+
+    def __init__(self, vocab_file, padding_side="left", clean_up_tokenization_spaces=False, encode_special_tokens=False,
+                 **kwargs):
+        self.name = "GLMTokenizer"
+
+        self.vocab_file = vocab_file
+        self.tokenizer = SPTokenizer(vocab_file)
+        self.special_tokens = {
+            "<bos>": self.tokenizer.bos_id,
+            "<eos>": self.tokenizer.eos_id,
+            "<pad>": self.tokenizer.pad_id
+        }
+        self.encode_special_tokens = encode_special_tokens
+        super().__init__(padding_side=padding_side, clean_up_tokenization_spaces=clean_up_tokenization_spaces,
+                         encode_special_tokens=encode_special_tokens,
+                         **kwargs)
+
+    def get_command(self, token):
+        if token in self.special_tokens:
+            return self.special_tokens[token]
+        assert token in self.tokenizer.special_tokens, f"{token} is not a special token for {self.name}"
+        return self.tokenizer.special_tokens[token]
+
+    @property
+    def unk_token(self) -> str:
+        return "<unk>"
+
+    @property
+    def pad_token(self) -> str:
+        return "<unk>"
+
+    @property
+    def pad_token_id(self):
+        return self.get_command("<pad>")
+
+    @property
+    def eos_token(self) -> str:
+        return "</s>"
+
+    @property
+    def eos_token_id(self):
+        return self.get_command("<eos>")
+
+    @property
+    def vocab_size(self):
+        return self.tokenizer.n_words
+
+    def get_vocab(self):
+        """ Returns vocab as a dict """
+        vocab = {self._convert_id_to_token(i): i for i in range(self.vocab_size)}
+        vocab.update(self.added_tokens_encoder)
+        return vocab
+
+    def _tokenize(self, text, **kwargs):
+        return self.tokenizer.tokenize(text, encode_special_tokens=self.encode_special_tokens)
+
+    def _convert_token_to_id(self, token):
+        """ Converts a token (str) in an id using the vocab. """
+        return self.tokenizer.convert_token_to_id(token)
+
+    def _convert_id_to_token(self, index):
+        """Converts an index (integer) in a token (str) using the vocab."""
+        return self.tokenizer.convert_id_to_token(index)
+
+    def convert_tokens_to_string(self, tokens: List[str]) -> str:
+        return self.tokenizer.decode_tokens(tokens)
+
+    def save_vocabulary(self, save_directory, filename_prefix=None):
+        """
+        Save the vocabulary and special tokens file to a directory.
+
+        Args:
+            save_directory (`str`):
+                The directory in which to save the vocabulary.
+            filename_prefix (`str`, *optional*):
+                An optional prefix to add to the named of the saved files.
+
+        Returns:
+            `Tuple(str)`: Paths to the files saved.
+        """
+        if os.path.isdir(save_directory):
+            vocab_file = os.path.join(
+                save_directory, self.vocab_files_names["vocab_file"]
+            )
+        else:
+            vocab_file = save_directory
+
+        with open(self.vocab_file, 'rb') as fin:
+            proto_str = fin.read()
+
+        with open(vocab_file, "wb") as writer:
+            writer.write(proto_str)
+
+        return (vocab_file,)
+
+    def get_prefix_tokens(self):
+        prefix_tokens = [self.get_command("[gMASK]"), self.get_command("sop")]
+        return prefix_tokens
+
+    def build_single_message(self, role, metadata, message):
+        assert role in ["system", "user", "assistant", "observation"], role
+        role_tokens = [self.get_command(f"<|{role}|>")] + self.tokenizer.encode(f"{metadata}\n")
+        message_tokens = self.tokenizer.encode(message)
+        tokens = role_tokens + message_tokens
+        return tokens
+
+    def build_chat_input(self, query, history=None, role="user"):
+        if history is None:
+            history = []
+        input_ids = []
+        for item in history:
+            content = item["content"]
+            if item["role"] == "system" and "tools" in item:
+                content = content + "\n" + json.dumps(item["tools"], indent=4, ensure_ascii=False)
+            input_ids.extend(self.build_single_message(item["role"], item.get("metadata", ""), content))
+        input_ids.extend(self.build_single_message(role, "", query))
+        input_ids.extend([self.get_command("<|assistant|>")])
+        return self.batch_encode_plus([input_ids], return_tensors="pt", is_split_into_words=True)
+
+    def build_inputs_with_special_tokens(
+            self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        """
+        Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and
+        adding special tokens. A BERT sequence has the following format:
+
+        - single sequence: `[CLS] X [SEP]`
+        - pair of sequences: `[CLS] A [SEP] B [SEP]`
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs to which the special tokens will be added.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+
+        Returns:
+            `List[int]`: List of [input IDs](../glossary#input-ids) with the appropriate special tokens.
+        """
+        prefix_tokens = self.get_prefix_tokens()
+        token_ids_0 = prefix_tokens + token_ids_0
+        if token_ids_1 is not None:
+            token_ids_0 = token_ids_0 + token_ids_1 + [self.get_command("<eos>")]
+        return token_ids_0
+
+    def _pad(
+            self,
+            encoded_inputs: Union[Dict[str, EncodedInput], BatchEncoding],
+            max_length: Optional[int] = None,
+            padding_strategy: PaddingStrategy = PaddingStrategy.DO_NOT_PAD,
+            pad_to_multiple_of: Optional[int] = None,
+            return_attention_mask: Optional[bool] = None,
+    ) -> dict:
+        """
+        Pad encoded inputs (on left/right and up to predefined length or max length in the batch)
+
+        Args:
+            encoded_inputs:
+                Dictionary of tokenized inputs (`List[int]`) or batch of tokenized inputs (`List[List[int]]`).
+            max_length: maximum length of the returned list and optionally padding length (see below).
+                Will truncate by taking into account the special tokens.
+            padding_strategy: PaddingStrategy to use for padding.
+
+                - PaddingStrategy.LONGEST Pad to the longest sequence in the batch
+                - PaddingStrategy.MAX_LENGTH: Pad to the max length (default)
+                - PaddingStrategy.DO_NOT_PAD: Do not pad
+                The tokenizer padding sides are defined in self.padding_side:
+
+                    - 'left': pads on the left of the sequences
+                    - 'right': pads on the right of the sequences
+            pad_to_multiple_of: (optional) Integer if set will pad the sequence to a multiple of the provided value.
+                This is especially useful to enable the use of Tensor Core on NVIDIA hardware with compute capability
+                `>= 7.5` (Volta).
+            return_attention_mask:
+                (optional) Set to False to avoid returning attention mask (default: set to model specifics)
+        """
+        # Load from model defaults
+        assert self.padding_side == "left"
+
+        required_input = encoded_inputs[self.model_input_names[0]]
+        seq_length = len(required_input)
+
+        if padding_strategy == PaddingStrategy.LONGEST:
+            max_length = len(required_input)
+
+        if max_length is not None and pad_to_multiple_of is not None and (max_length % pad_to_multiple_of != 0):
+            max_length = ((max_length // pad_to_multiple_of) + 1) * pad_to_multiple_of
+
+        needs_to_be_padded = padding_strategy != PaddingStrategy.DO_NOT_PAD and len(required_input) != max_length
+
+        # Initialize attention mask if not present.
+        if "attention_mask" not in encoded_inputs:
+            encoded_inputs["attention_mask"] = [1] * seq_length
+
+        if "position_ids" not in encoded_inputs:
+            encoded_inputs["position_ids"] = list(range(seq_length))
+
+        if needs_to_be_padded:
+            difference = max_length - len(required_input)
+
+            if "attention_mask" in encoded_inputs:
+                encoded_inputs["attention_mask"] = [0] * difference + encoded_inputs["attention_mask"]
+            if "position_ids" in encoded_inputs:
+                encoded_inputs["position_ids"] = [0] * difference + encoded_inputs["position_ids"]
+            encoded_inputs[self.model_input_names[0]] = [self.pad_token_id] * difference + required_input
+
+        return encoded_inputs
diff --git a/chatglm3-6b/tokenizer.model b/chatglm3-6b/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..8a8007697b7cc3d3868dcffbbebf8c1f2bd690ba
--- /dev/null
+++ b/chatglm3-6b/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7dc4c393423b76e4373e5157ddc34803a0189ba96b21ddbb40269d31468a6f2
+size 1018370
diff --git a/chatglm3-6b/tokenizer_config.json b/chatglm3-6b/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..16882eb6bc81e8abf79166612613fefcfb9d3d25
--- /dev/null
+++ b/chatglm3-6b/tokenizer_config.json
@@ -0,0 +1,12 @@
+{
+  "name_or_path": "THUDM/chatglm3-6b",
+  "remove_space": false,
+  "do_lower_case": false,
+  "tokenizer_class": "ChatGLMTokenizer",
+  "auto_map": {
+    "AutoTokenizer": [
+      "tokenization_chatglm.ChatGLMTokenizer",
+      null
+      ]
+  }
+}
diff --git a/composite_demo/.streamlit/config.toml b/composite_demo/.streamlit/config.toml
new file mode 100644
index 0000000000000000000000000000000000000000..30bc5ccec89368643b5db2dd34202a52d4d10382
--- /dev/null
+++ b/composite_demo/.streamlit/config.toml
@@ -0,0 +1,2 @@
+[theme]
+font = "monospace"
\ No newline at end of file
diff --git a/composite_demo/README.md b/composite_demo/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..428dd6620539ac90bd20f2fd1d264df391b5425d
--- /dev/null
+++ b/composite_demo/README.md
@@ -0,0 +1,85 @@
+# ChatGLM3 Web Demo
+
+![Demo webpage](assets/demo.png)
+
+## 安装
+
+我们建议通过 [Conda](https://docs.conda.io/en/latest/) 进行环境管理。
+
+执行以下命令新建一个 conda 环境并安装所需依赖：
+
+```bash
+conda create -n chatglm3-demo python=3.10
+conda activate chatglm3-demo
+pip install -r requirements.txt
+```
+
+请注意，本项目需要 Python 3.10 或更高版本。
+
+此外，使用 Code Interpreter 还需要安装 Jupyter 内核：
+
+```bash
+ipython kernel install --name chatglm3-demo --user
+```
+
+## 运行
+
+运行以下命令在本地加载模型并启动 demo：
+
+```bash
+streamlit run main.py
+```
+
+之后即可从命令行中看到 demo 的地址，点击即可访问。初次访问需要下载并加载模型，可能需要花费一定时间。
+
+如果已经在本地下载了模型，可以通过 `export MODEL_PATH=/path/to/model` 来指定从本地加载模型。如果需要自定义 Jupyter 内核，可以通过 `export IPYKERNEL=<kernel_name>` 来指定。
+
+## 使用
+
+ChatGLM3 Demo 拥有三种模式：
+
+- Chat: 对话模式，在此模式下可以与模型进行对话。
+- Tool: 工具模式，模型除了对话外，还可以通过工具进行其他操作。
+- Code Interpreter: 代码解释器模式，模型可以在一个 Jupyter 环境中执行代码并获取结果，以完成复杂任务。
+
+### 对话模式
+
+对话模式下，用户可以直接在侧边栏修改 top_p, temperature, System Prompt 等参数来调整模型的行为。例如
+
+![The model responses following system prompt](assets/emojis.png)
+
+### 工具模式
+
+可以通过在 `tool_registry.py` 中注册新的工具来增强模型的能力。只需要使用 `@register_tool` 装饰函数即可完成注册。对于工具声明，函数名称即为工具的名称，函数 docstring 即为工具的说明；对于工具的参数，使用 `Annotated[typ: type, description: str, required: bool]` 标注参数的类型、描述和是否必须。
+
+例如，`get_weather` 工具的注册如下：
+
+```python
+@register_tool
+def get_weather(
+    city_name: Annotated[str, 'The name of the city to be queried', True],
+) -> str:
+    """
+    Get the weather for `city_name` in the following week
+    """
+    ...
+```
+
+![The model uses tool to query the weather of pairs.](assets/tool.png)
+
+此外，你也可以在页面中通过 `Manual mode` 进入手动模式，在这一模式下你可以通过 YAML 来直接指定工具列表，但你需要手动将工具的输出反馈给模型。
+
+### 代码解释器模式
+
+由于拥有代码执行环境，此模式下的模型能够执行更为复杂的任务，例如绘制图表、执行符号运算等等。模型会根据对任务完成情况的理解自动地连续执行多个代码块，直到任务完成。因此，在这一模式下，你只需要指明希望模型执行的任务即可。
+
+例如，我们可以让 ChatGLM3 画一个爱心：
+
+![The code interpreter draws a heart according to the user's instructions.](assets/heart.png)
+
+### 额外技巧
+
+- 在模型生成文本时，可以通过页面右上角的 `Stop` 按钮进行打断。
+- 刷新页面即可清空对话记录。
+
+# Enjoy!
\ No newline at end of file
diff --git a/composite_demo/README_en.md b/composite_demo/README_en.md
new file mode 100644
index 0000000000000000000000000000000000000000..67d6d5a5e04f0c271703ff2939a48f99fbe1db60
--- /dev/null
+++ b/composite_demo/README_en.md
@@ -0,0 +1,85 @@
+# ChatGLM3 Web Demo
+
+![Demo webpage](assets/demo.png)
+
+## Installation
+
+We recommend managing environments through [Conda](https://docs.conda.io/en/latest/).
+
+Execute the following commands to create a new conda environment and install the necessary dependencies:
+
+```bash
+conda create -n chatglm3-demo python=3.10
+conda activate chatglm3-demo
+pip install -r requirements.txt
+```
+
+Please note that this project requires Python 3.10 or higher.
+
+Additionally, installing the Jupyter kernel is required for using the Code Interpreter:
+
+```bash
+ipython kernel install --name chatglm3-demo --user
+```
+
+## Execution
+
+Run the following command to load the model locally and start the demo:
+
+```bash
+streamlit run main.py
+```
+
+Afterward, the address of the demo can be seen from the command line; click to access. The first visit requires the download and loading of the model, which may take some time.
+
+If the model has already been downloaded locally, you can specify to load the model locally through `export MODEL_PATH=/path/to/model`. If you need to customize the Jupyter kernel, you can specify it through `export IPYKERNEL=<kernel_name>`.
+
+## Usage
+
+ChatGLM3 Demo has three modes:
+
+- Chat: Dialogue mode, where you can interact with the model.
+- Tool: Tool mode, where the model, in addition to dialogue, can perform other operations through tools.
+- Code Interpreter: Code interpreter mode, where the model can execute code in a Jupyter environment and obtain results to complete complex tasks.
+
+### Dialogue Mode
+
+In dialogue mode, users can directly modify parameters such as top_p, temperature, System Prompt in the sidebar to adjust the behavior of the model. For example,
+
+![The model responses following system prompt](assets/emojis.png)
+
+### Tool Mode
+
+You can enhance the model's capabilities by registering new tools in `tool_registry.py`. Just use the `@register_tool` decorator to complete the registration. For tool declarations, the function name is the name of the tool, and the function docstring is the description of the tool; for tool parameters, use `Annotated[typ: type, description: str, required: bool]` to annotate the type, description, and whether it is necessary of the parameters.
+
+For example, the registration of the `get_weather` tool is as follows:
+
+```python
+@register_tool
+def get_weather(
+    city_name: Annotated[str, 'The name of the city to be queried', True],
+) -> str:
+    """
+    Get the weather for `city_name` in the following week
+    """
+    ...
+```
+
+![The model uses tool to query the weather of pairs.](assets/tool.png)
+
+Additionally, you can enter the manual mode through `Manual mode` on the page. In this mode, you can directly specify the tool list through YAML, but you need to manually feed back the tool's output to the model.
+
+### Code Interpreter Mode
+
+Due to having a code execution environment, the model in this mode can perform more complex tasks, such as drawing charts, performing symbolic operations, etc. The model will automatically execute multiple code blocks in succession based on its understanding of the task completion status until the task is completed. Therefore, in this mode, you only need to specify the task you want the model to perform.
+
+For example, we can ask ChatGLM3 to draw a heart:
+
+![The code interpreter draws a heart according to the user's instructions.](assets/heart.png)
+
+### Additional Tips
+
+- While the model is generating text, it can be interrupted by the `Stop` button at the top right corner of the page.
+- Refreshing the page will clear the dialogue history.
+
+# Enjoy!
\ No newline at end of file
diff --git a/composite_demo/assets/demo.png b/composite_demo/assets/demo.png
new file mode 100644
index 0000000000000000000000000000000000000000..5d2b10f2d626bca685adfd7a474f25e29e767bdd
Binary files /dev/null and b/composite_demo/assets/demo.png differ
diff --git a/composite_demo/assets/emojis.png b/composite_demo/assets/emojis.png
new file mode 100644
index 0000000000000000000000000000000000000000..9bc293d9f30d3114ad24b22b88105d6ec94775aa
Binary files /dev/null and b/composite_demo/assets/emojis.png differ
diff --git a/composite_demo/assets/heart.png b/composite_demo/assets/heart.png
new file mode 100644
index 0000000000000000000000000000000000000000..0ee1462c14d2aab783ad66a1eccf451308514f2e
Binary files /dev/null and b/composite_demo/assets/heart.png differ
diff --git a/composite_demo/assets/tool.png b/composite_demo/assets/tool.png
new file mode 100644
index 0000000000000000000000000000000000000000..29bca13e8c93d57631d2a2e6c2ea68dd2699d8a9
Binary files /dev/null and b/composite_demo/assets/tool.png differ
diff --git a/composite_demo/client.py b/composite_demo/client.py
new file mode 100644
index 0000000000000000000000000000000000000000..a6873d171be77a61fad5ddb425bc46bc901ad5af
--- /dev/null
+++ b/composite_demo/client.py
@@ -0,0 +1,197 @@
+from __future__ import annotations
+
+from collections.abc import Iterable
+import os
+from typing import Any, Protocol
+
+from huggingface_hub.inference._text_generation import TextGenerationStreamResponse, Token
+import streamlit as st
+import torch
+from transformers import AutoModel, AutoTokenizer, AutoConfig
+
+from conversation import Conversation
+
+TOOL_PROMPT = 'Answer the following questions as best as you can. You have access to the following tools:'
+
+MODEL_PATH = os.environ.get('MODEL_PATH', 'THUDM/chatglm3-6b')
+PT_PATH = os.environ.get('PT_PATH', None)
+TOKENIZER_PATH = os.environ.get("TOKENIZER_PATH", MODEL_PATH)
+DEVICE = 'cuda' if torch.cuda.is_available() else 'cpu'
+
+# for Mac Computer like M1
+# You Need Use Pytorch compiled with Metal
+# DEVICE = 'mps'
+
+# for AMD gpu likes MI100 (Not Official Steady Support yet)
+# You Need Use Pytorch compiled with ROCm
+# DEVICE = 'cuda'
+
+# for Intel gpu likes A770 (Not Official Steady Support yet)
+# You Need Use Pytorch compiled with oneDNN and install intel-extension-for-pytorch
+# import intel_extension_for_pytorch as ipex
+# DEVICE = 'xpu'
+
+# for Moore Threads gpu like MTT S80 (Not Official Steady Support yet)
+# You Need Use Pytorch compiled with Musa
+# DEVICE = 'musa'
+
+
+@st.cache_resource
+def get_client() -> Client:
+    client = HFClient(MODEL_PATH, TOKENIZER_PATH, PT_PATH, DEVICE)
+    return client
+
+
+class Client(Protocol):
+    def generate_stream(self,
+                        system: str | None,
+                        tools: list[dict] | None,
+                        history: list[Conversation],
+                        **parameters: Any
+                        ) -> Iterable[TextGenerationStreamResponse]:
+        ...
+
+
+def stream_chat(self, tokenizer, query: str, history: list[tuple[str, str]] = None, role: str = "user",
+                past_key_values=None, max_length: int = 8192, do_sample=True, top_p=0.8, temperature=0.8,
+                repetition_penalty=1.0, length_penalty=1.0, num_beams=1,
+                logits_processor=None, return_past_key_values=False, **kwargs):
+    from transformers.generation.logits_process import LogitsProcessor
+    from transformers.generation.utils import LogitsProcessorList
+
+    class InvalidScoreLogitsProcessor(LogitsProcessor):
+        def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTensor) -> torch.FloatTensor:
+            if torch.isnan(scores).any() or torch.isinf(scores).any():
+                scores.zero_()
+                scores[..., 5] = 5e4
+            return scores
+
+    if history is None:
+        history = []
+    if logits_processor is None:
+        logits_processor = LogitsProcessorList()
+    logits_processor.append(InvalidScoreLogitsProcessor())
+    eos_token_id = [tokenizer.eos_token_id, tokenizer.get_command("<|user|>"),
+                    tokenizer.get_command("<|observation|>")]
+    gen_kwargs = {"max_length": max_length,
+                  "do_sample": do_sample,
+                  "top_p": top_p,
+                  "temperature": temperature,
+                  "logits_processor": logits_processor,
+                  "repetition_penalty": repetition_penalty,
+                  "length_penalty": length_penalty,
+                  "num_beams": num_beams,
+                  **kwargs
+                  }
+
+    print(gen_kwargs)
+    if past_key_values is None:
+        inputs = tokenizer.build_chat_input(query, history=history, role=role)
+    else:
+        inputs = tokenizer.build_chat_input(query, role=role)
+    inputs = inputs.to(self.device)
+    if past_key_values is not None:
+        past_length = past_key_values[0][0].shape[0]
+        if self.transformer.pre_seq_len is not None:
+            past_length -= self.transformer.pre_seq_len
+        inputs.position_ids += past_length
+        attention_mask = inputs.attention_mask
+        attention_mask = torch.cat((attention_mask.new_ones(1, past_length), attention_mask), dim=1)
+        inputs['attention_mask'] = attention_mask
+    history.append({"role": role, "content": query})
+    print("input_shape>", inputs['input_ids'].shape)
+
+    input_sequence_length = inputs['input_ids'].shape[1]
+
+    if max_length < input_sequence_length <= self.config.seq_length:
+        yield "Current input sequence length {} exceeds sequence length set in generation parameters {}. The maximum model sequence length is {}. You may adjust the generation parameter to enable longer chat history.".format(
+            input_sequence_length, max_length, self.config.seq_length
+        ), history
+        return
+
+    if input_sequence_length > self.config.seq_length:
+        yield "Current input sequence length {} exceeds maximum model sequence length {}. Unable to generate tokens.".format(
+            input_sequence_length, self.config.seq_length
+        ), history
+        return
+
+    for outputs in self.stream_generate(**inputs, past_key_values=past_key_values,
+                                        eos_token_id=eos_token_id, return_past_key_values=return_past_key_values,
+                                        **gen_kwargs):
+        if return_past_key_values:
+            outputs, past_key_values = outputs
+        outputs = outputs.tolist()[0][len(inputs["input_ids"][0]):]
+        response = tokenizer.decode(outputs)
+        if response and response[-1] != "�":
+            new_history = history
+            if return_past_key_values:
+                yield response, new_history, past_key_values
+            else:
+                yield response, new_history
+
+
+class HFClient(Client):
+    def __init__(self, model_path: str, tokenizer_path: str, pt_checkpoint: str | None = None, DEVICE = 'cpu'):
+        self.model_path = model_path
+        self.tokenizer = AutoTokenizer.from_pretrained(tokenizer_path, trust_remote_code=True)
+
+        if pt_checkpoint is not None:
+            config = AutoConfig.from_pretrained(model_path, trust_remote_code=True, pre_seq_len=128)
+            self.model = AutoModel.from_pretrained(model_path, trust_remote_code=True, config=config)
+            prefix_state_dict = torch.load(os.path.join(pt_checkpoint, "pytorch_model.bin"))
+            new_prefix_state_dict = {}
+            for k, v in prefix_state_dict.items():
+                if k.startswith("transformer.prefix_encoder."):
+                    new_prefix_state_dict[k[len("transformer.prefix_encoder."):]] = v
+            print("Loaded from pt checkpoints", new_prefix_state_dict.keys())
+            self.model.transformer.prefix_encoder.load_state_dict(new_prefix_state_dict)
+        else:
+            self.model = AutoModel.from_pretrained(model_path, trust_remote_code=True)
+
+        self.model = self.model.to(DEVICE).eval() if 'cuda' in DEVICE else self.model.float().to(DEVICE).eval()
+
+
+    def generate_stream(self,
+                        system: str | None,
+                        tools: list[dict] | None,
+                        history: list[Conversation],
+                        **parameters: Any
+                        ) -> Iterable[TextGenerationStreamResponse]:
+        chat_history = [{
+            'role': 'system',
+            'content': system if not tools else TOOL_PROMPT,
+        }]
+
+        if tools:
+            chat_history[0]['tools'] = tools
+
+        for conversation in history[:-1]:
+            chat_history.append({
+                'role': str(conversation.role).removeprefix('<|').removesuffix('|>'),
+                'content': conversation.content,
+            })
+
+        query = history[-1].content
+        role = str(history[-1].role).removeprefix('<|').removesuffix('|>')
+
+        text = ''
+
+        for new_text, _ in stream_chat(self.model,
+                                       self.tokenizer,
+                                       query,
+                                       chat_history,
+                                       role,
+                                       **parameters,
+                                       ):
+            word = new_text.removeprefix(text)
+            word_stripped = word.strip()
+            text = new_text
+            yield TextGenerationStreamResponse(
+                generated_text=text,
+                token=Token(
+                    id=0,
+                    logprob=0,
+                    text=word,
+                    special=word_stripped.startswith('<|') and word_stripped.endswith('|>'),
+                )
+            )
diff --git a/composite_demo/conversation.py b/composite_demo/conversation.py
new file mode 100644
index 0000000000000000000000000000000000000000..1ac13e774775dfad8e18e728e8b33ca2a40b8f65
--- /dev/null
+++ b/composite_demo/conversation.py
@@ -0,0 +1,119 @@
+from dataclasses import dataclass
+from enum import auto, Enum
+import json
+
+from PIL.Image import Image
+import streamlit as st
+from streamlit.delta_generator import DeltaGenerator
+
+TOOL_PROMPT = 'Answer the following questions as best as you can. You have access to the following tools:\n'
+
+class Role(Enum):
+    SYSTEM = auto()
+    USER = auto()
+    ASSISTANT = auto()
+    TOOL = auto()
+    INTERPRETER = auto()
+    OBSERVATION = auto()
+
+    def __str__(self):
+        match self:
+            case Role.SYSTEM:
+                return "<|system|>"
+            case Role.USER:
+                return "<|user|>"
+            case Role.ASSISTANT | Role.TOOL | Role.INTERPRETER:
+                return "<|assistant|>"
+            case Role.OBSERVATION:
+                return "<|observation|>"
+            
+    # Get the message block for the given role
+    def get_message(self):
+        # Compare by value here, because the enum object in the session state
+        # is not the same as the enum cases here, due to streamlit's rerunning
+        # behavior.
+        match self.value:
+            case Role.SYSTEM.value:
+                return
+            case Role.USER.value:
+                return st.chat_message(name="user", avatar="user")
+            case Role.ASSISTANT.value:
+                return st.chat_message(name="assistant", avatar="assistant")
+            case Role.TOOL.value:
+                return st.chat_message(name="tool", avatar="assistant")
+            case Role.INTERPRETER.value:
+                return st.chat_message(name="interpreter", avatar="assistant")
+            case Role.OBSERVATION.value:
+                return st.chat_message(name="observation", avatar="user")
+            case _:
+                st.error(f'Unexpected role: {self}')
+
+@dataclass
+class Conversation:
+    role: Role
+    content: str
+    tool: str | None = None
+    image: Image | None = None
+
+    def __str__(self) -> str:
+        print(self.role, self.content, self.tool)
+        match self.role:
+            case Role.SYSTEM | Role.USER | Role.ASSISTANT | Role.OBSERVATION:
+                return f'{self.role}\n{self.content}'
+            case Role.TOOL:
+                return f'{self.role}{self.tool}\n{self.content}'
+            case Role.INTERPRETER:
+                return f'{self.role}interpreter\n{self.content}'
+
+    # Human readable format
+    def get_text(self) -> str:
+        text = postprocess_text(self.content)
+        match self.role.value:
+            case Role.TOOL.value:
+                text = f'Calling tool `{self.tool}`:\n{text}'
+            case Role.INTERPRETER.value:
+                text = f'{text}'
+            case Role.OBSERVATION.value:
+                text = f'Observation:\n```\n{text}\n```'
+        return text
+    
+    # Display as a markdown block
+    def show(self, placeholder: DeltaGenerator | None=None) -> str:
+        if placeholder:
+            message = placeholder
+        else:
+            message = self.role.get_message()
+        if self.image:
+            message.image(self.image)
+        else:
+            text = self.get_text()
+            message.markdown(text)
+
+def preprocess_text(
+    system: str | None,
+    tools: list[dict] | None,
+    history: list[Conversation],
+) -> str:
+    if tools:
+        tools = json.dumps(tools, indent=4, ensure_ascii=False)
+
+    prompt = f"{Role.SYSTEM}\n"
+    prompt += system if not tools else TOOL_PROMPT
+    if tools:
+        tools = json.loads(tools)
+        prompt += json.dumps(tools, ensure_ascii=False)
+    for conversation in history:
+        prompt += f'{conversation}'
+    prompt += f'{Role.ASSISTANT}\n'
+    return prompt
+
+def postprocess_text(text: str) -> str:
+    text = text.replace("\(", "$")
+    text = text.replace("\)", "$")
+    text = text.replace("\[", "$$")
+    text = text.replace("\]", "$$")
+    text = text.replace("<|assistant|>", "")
+    text = text.replace("<|observation|>", "")
+    text = text.replace("<|system|>", "")
+    text = text.replace("<|user|>", "")
+    return text.strip()
\ No newline at end of file
diff --git a/composite_demo/demo_chat.py b/composite_demo/demo_chat.py
new file mode 100644
index 0000000000000000000000000000000000000000..e8b8d99a1af6889a64b8573e3e69687f281af05d
--- /dev/null
+++ b/composite_demo/demo_chat.py
@@ -0,0 +1,80 @@
+import streamlit as st
+from streamlit.delta_generator import DeltaGenerator
+
+from client import get_client
+from conversation import postprocess_text, preprocess_text, Conversation, Role
+
+MAX_LENGTH = 8192
+
+client = get_client()
+
+
+# Append a conversation into history, while show it in a new markdown block
+def append_conversation(
+        conversation: Conversation,
+        history: list[Conversation],
+        placeholder: DeltaGenerator | None = None,
+) -> None:
+    history.append(conversation)
+    conversation.show(placeholder)
+
+
+def main(top_p: float, temperature: float, system_prompt: str, prompt_text: str, repetition_penalty: float):
+    placeholder = st.empty()
+    with placeholder.container():
+        if 'chat_history' not in st.session_state:
+            st.session_state.chat_history = []
+
+        history: list[Conversation] = st.session_state.chat_history
+
+        for conversation in history:
+            conversation.show()
+
+    if prompt_text:
+        prompt_text = prompt_text.strip()
+        append_conversation(Conversation(Role.USER, prompt_text), history)
+
+        input_text = preprocess_text(
+            system_prompt,
+            tools=None,
+            history=history,
+        )
+        print("=== Input:")
+        print(input_text)
+        print("=== History:")
+        print(history)
+
+        placeholder = st.empty()
+        message_placeholder = placeholder.chat_message(name="assistant", avatar="assistant")
+        markdown_placeholder = message_placeholder.empty()
+
+        output_text = ''
+        for response in client.generate_stream(
+                system_prompt,
+                tools=None,
+                history=history,
+                do_sample=True,
+                max_length=MAX_LENGTH,
+                temperature=temperature,
+                top_p=top_p,
+                stop_sequences=[str(Role.USER)],
+                repetition_penalty=repetition_penalty,
+        ):
+            token = response.token
+            if response.token.special:
+                print("=== Output:")
+                print(output_text)
+
+                match token.text.strip():
+                    case '<|user|>':
+                        break
+                    case _:
+                        st.error(f'Unexpected special token: {token.text.strip()}')
+                        break
+            output_text += response.token.text
+            markdown_placeholder.markdown(postprocess_text(output_text + '▌'))
+
+        append_conversation(Conversation(
+            Role.ASSISTANT,
+            postprocess_text(output_text),
+        ), history, markdown_placeholder)
diff --git a/composite_demo/demo_ci.py b/composite_demo/demo_ci.py
new file mode 100644
index 0000000000000000000000000000000000000000..4edf29e69080bc38c6b5068d7d5f37d248ce2858
--- /dev/null
+++ b/composite_demo/demo_ci.py
@@ -0,0 +1,329 @@
+import base64
+from io import BytesIO
+import os
+from pprint import pprint
+import queue
+import re
+from subprocess import PIPE
+
+import jupyter_client
+from PIL import Image
+import streamlit as st
+from streamlit.delta_generator import DeltaGenerator
+
+from client import get_client
+from conversation import postprocess_text, preprocess_text, Conversation, Role
+
+IPYKERNEL = os.environ.get('IPYKERNEL', 'chatglm3-demo')
+
+SYSTEM_PROMPT = '你是一位智能AI助手，你叫ChatGLM，你连接着一台电脑，但请注意不能联网。在使用Python解决任务时，你可以运行代码并得到结果，如果运行结果有错误，你需要尽可能对代码进行改进。你可以处理用户上传到电脑上的文件，文件默认存储路径是/mnt/data/。'
+
+MAX_LENGTH = 8192
+TRUNCATE_LENGTH = 1024
+
+client = get_client()
+
+class CodeKernel(object):
+    def __init__(self,
+                 kernel_name='kernel',
+                 kernel_id=None,
+                 kernel_config_path="",
+                 python_path=None,
+                 ipython_path=None,
+                 init_file_path="./startup.py",
+                 verbose=1):
+
+        self.kernel_name = kernel_name
+        self.kernel_id = kernel_id
+        self.kernel_config_path = kernel_config_path
+        self.python_path = python_path
+        self.ipython_path = ipython_path
+        self.init_file_path = init_file_path
+        self.verbose = verbose
+        
+        if python_path is None and ipython_path is None:
+            env = None
+        else:
+            env = {"PATH": self.python_path + ":$PATH", "PYTHONPATH": self.python_path}
+
+        # Initialize the backend kernel
+        self.kernel_manager = jupyter_client.KernelManager(kernel_name=IPYKERNEL, 
+                                                           connection_file=self.kernel_config_path,
+                                                           exec_files=[self.init_file_path],
+                                                           env=env)
+        if self.kernel_config_path:
+            self.kernel_manager.load_connection_file()
+            self.kernel_manager.start_kernel(stdout=PIPE, stderr=PIPE)
+            print("Backend kernel started with the configuration: {}".format(
+                self.kernel_config_path))
+        else:
+            self.kernel_manager.start_kernel(stdout=PIPE, stderr=PIPE)
+            print("Backend kernel started with the configuration: {}".format(
+                self.kernel_manager.connection_file))
+
+        if verbose:
+            pprint(self.kernel_manager.get_connection_info())
+
+        # Initialize the code kernel
+        self.kernel = self.kernel_manager.blocking_client()
+        # self.kernel.load_connection_file()
+        self.kernel.start_channels()
+        print("Code kernel started.")
+
+    def execute(self, code):
+        self.kernel.execute(code)
+        try:
+            shell_msg = self.kernel.get_shell_msg(timeout=30)
+            io_msg_content = self.kernel.get_iopub_msg(timeout=30)['content']
+            while True:
+                msg_out = io_msg_content
+                ### Poll the message
+                try:
+                    io_msg_content = self.kernel.get_iopub_msg(timeout=30)['content']
+                    if 'execution_state' in io_msg_content and io_msg_content['execution_state'] == 'idle':
+                        break
+                except queue.Empty:
+                    break
+            
+            return shell_msg, msg_out
+        except Exception as e:
+            print(e)
+            return None
+
+    def execute_interactive(self, code, verbose=False):
+        shell_msg = self.kernel.execute_interactive(code)
+        if shell_msg is queue.Empty:
+            if verbose:
+                print("Timeout waiting for shell message.")
+        self.check_msg(shell_msg, verbose=verbose)
+
+        return shell_msg
+
+    def inspect(self, code, verbose=False):
+        msg_id = self.kernel.inspect(code)
+        shell_msg = self.kernel.get_shell_msg(timeout=30)
+        if shell_msg is queue.Empty:
+            if verbose:
+                print("Timeout waiting for shell message.")
+        self.check_msg(shell_msg, verbose=verbose)
+
+        return shell_msg
+
+    def get_error_msg(self, msg, verbose=False) -> str | None:
+        if msg['content']['status'] == 'error':
+            try:
+                error_msg = msg['content']['traceback']
+            except:
+                try:
+                    error_msg = msg['content']['traceback'][-1].strip()
+                except:
+                    error_msg = "Traceback Error"
+            if verbose:
+                print("Error: ", error_msg)
+            return error_msg
+        return None
+
+    def check_msg(self, msg, verbose=False):
+        status = msg['content']['status']
+        if status == 'ok':
+            if verbose:
+                print("Execution succeeded.")
+        elif status == 'error':
+            for line in msg['content']['traceback']:
+                if verbose:
+                    print(line)
+
+    def shutdown(self):
+        # Shutdown the backend kernel
+        self.kernel_manager.shutdown_kernel()
+        print("Backend kernel shutdown.")
+        # Shutdown the code kernel
+        self.kernel.shutdown()
+        print("Code kernel shutdown.")
+
+    def restart(self):
+        # Restart the backend kernel
+        self.kernel_manager.restart_kernel()
+        # print("Backend kernel restarted.")
+
+    def interrupt(self):
+        # Interrupt the backend kernel
+        self.kernel_manager.interrupt_kernel()
+        # print("Backend kernel interrupted.")
+
+    def is_alive(self):
+        return self.kernel.is_alive()
+    
+def b64_2_img(data):
+    buff = BytesIO(base64.b64decode(data))
+    return Image.open(buff)
+
+def clean_ansi_codes(input_string):
+    ansi_escape = re.compile(r'(\x9B|\x1B\[|\u001b\[)[0-?]*[ -/]*[@-~]')
+    return ansi_escape.sub('', input_string)
+    
+def execute(code, kernel: CodeKernel) -> tuple[str, str | Image.Image]:
+    res = ""
+    res_type = None
+    code = code.replace("<|observation|>", "")
+    code = code.replace("<|assistant|>interpreter", "")
+    code = code.replace("<|assistant|>", "")
+    code = code.replace("<|user|>", "")
+    code = code.replace("<|system|>", "")
+    msg, output = kernel.execute(code)
+    
+    if msg['metadata']['status'] == "timeout":
+        return res_type, 'Timed out'
+    elif msg['metadata']['status'] == 'error':
+        return res_type, clean_ansi_codes('\n'.join(kernel.get_error_msg(msg, verbose=True)))
+    
+    if 'text' in output:
+        res_type = "text"
+        res = output['text']
+    elif 'data' in output:
+        for key in output['data']:
+            if 'text/plain' in key:
+                res_type = "text"
+                res = output['data'][key]
+            elif 'image/png' in key:
+                res_type = "image"
+                res = output['data'][key]
+                break
+
+    if res_type == "image":
+        return res_type, b64_2_img(res)
+    elif res_type == "text" or res_type == "traceback":
+        res = res
+    
+    return res_type, res
+
+@st.cache_resource
+def get_kernel():
+    kernel = CodeKernel()
+    return kernel
+
+def extract_code(text: str) -> str:
+    pattern = r'```([^\n]*)\n(.*?)```'
+    matches = re.findall(pattern, text, re.DOTALL)
+    return matches[-1][1]
+
+# Append a conversation into history, while show it in a new markdown block
+def append_conversation(
+    conversation: Conversation,
+    history: list[Conversation],
+    placeholder: DeltaGenerator | None=None,
+) -> None:
+    history.append(conversation)
+    conversation.show(placeholder)
+
+def main(top_p: float, temperature: float, prompt_text: str, repetition_penalty: float):
+    if 'ci_history' not in st.session_state:
+        st.session_state.ci_history = []
+
+    history: list[Conversation] = st.session_state.ci_history
+
+    for conversation in history:
+        conversation.show()
+
+    if prompt_text:
+        prompt_text = prompt_text.strip()
+        role = Role.USER
+        append_conversation(Conversation(role, prompt_text), history)
+
+        input_text = preprocess_text(
+            SYSTEM_PROMPT,
+            None,
+            history,
+        )
+        print("=== Input:")
+        print(input_text)
+        print("=== History:")
+        print(history)
+
+        placeholder = st.container()
+        message_placeholder = placeholder.chat_message(name="assistant", avatar="assistant")
+        markdown_placeholder = message_placeholder.empty()
+
+        for _ in range(5):
+            output_text = ''
+            for response in client.generate_stream(
+                system=SYSTEM_PROMPT,
+                tools=None,
+                history=history,
+                do_sample=True,
+                max_length=MAX_LENGTH,
+                temperature=temperature,
+                top_p=top_p,
+                stop_sequences=[str(r) for r in (Role.USER, Role.OBSERVATION)],
+                repetition_penalty=repetition_penalty,
+            ):
+                token = response.token
+                if response.token.special:
+                    print("=== Output:")
+                    print(output_text)
+
+                    match token.text.strip():
+                        case '<|user|>':
+                            append_conversation(Conversation(
+                                Role.ASSISTANT,
+                                postprocess_text(output_text),
+                            ), history, markdown_placeholder)
+                            return
+                        # Initiate tool call
+                        case '<|assistant|>':
+                            append_conversation(Conversation(
+                                Role.ASSISTANT,
+                                postprocess_text(output_text),
+                            ), history, markdown_placeholder)
+                            message_placeholder = placeholder.chat_message(name="interpreter", avatar="assistant")
+                            markdown_placeholder = message_placeholder.empty()
+                            output_text = ''
+                            continue
+                        case '<|observation|>':
+                            code = extract_code(output_text)
+                            print("Code:", code)
+
+                            display_text = output_text.split('interpreter')[-1].strip()
+                            append_conversation(Conversation(
+                                Role.INTERPRETER,
+                                postprocess_text(display_text),
+                            ), history, markdown_placeholder)
+                            message_placeholder = placeholder.chat_message(name="observation", avatar="user")
+                            markdown_placeholder = message_placeholder.empty()
+                            output_text = ''
+                            
+                            with markdown_placeholder:
+                                with st.spinner('Executing code...'):
+                                    try:
+                                        res_type, res = execute(code, get_kernel())
+                                    except Exception as e:
+                                        st.error(f'Error when executing code: {e}')
+                                        return
+                            print("Received:", res_type, res)
+
+                            if res_type == 'text' and len(res) > TRUNCATE_LENGTH:
+                                res = res[:TRUNCATE_LENGTH] + ' [TRUNCATED]'
+
+                            append_conversation(Conversation(
+                                Role.OBSERVATION,
+                                '[Image]' if res_type == 'image' else postprocess_text(res),
+                                tool=None,
+                                image=res if res_type == 'image' else None,
+                            ), history, markdown_placeholder)
+                            message_placeholder = placeholder.chat_message(name="assistant", avatar="assistant")
+                            markdown_placeholder = message_placeholder.empty()
+                            output_text = ''
+                            break
+                        case _:
+                            st.error(f'Unexpected special token: {token.text.strip()}')
+                            break
+                output_text += response.token.text
+                display_text = output_text.split('interpreter')[-1].strip()
+                markdown_placeholder.markdown(postprocess_text(display_text + '▌'))
+            else:
+                append_conversation(Conversation(
+                    Role.ASSISTANT,
+                    postprocess_text(output_text),
+                ), history, markdown_placeholder)
+                return
+
diff --git a/composite_demo/demo_tool.py b/composite_demo/demo_tool.py
new file mode 100644
index 0000000000000000000000000000000000000000..68a1b886752b41845dc27f7a0a3e6733aa5d2126
--- /dev/null
+++ b/composite_demo/demo_tool.py
@@ -0,0 +1,192 @@
+import re
+import yaml
+from yaml import YAMLError
+
+import streamlit as st
+from streamlit.delta_generator import DeltaGenerator
+
+from client import get_client
+from conversation import postprocess_text, preprocess_text, Conversation, Role
+from tool_registry import dispatch_tool, get_tools
+
+MAX_LENGTH = 8192
+TRUNCATE_LENGTH = 1024
+
+EXAMPLE_TOOL = {
+    "name": "get_current_weather",
+    "description": "Get the current weather in a given location",
+    "parameters": {
+        "type": "object",
+        "properties": {
+            "location": {
+                "type": "string",
+                "description": "The city and state, e.g. San Francisco, CA",
+            },
+            "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]},
+        },
+        "required": ["location"],
+    }
+}
+
+client = get_client()
+
+def tool_call(*args, **kwargs) -> dict:
+    print("=== Tool call:")
+    print(args)
+    print(kwargs)
+    st.session_state.calling_tool = True
+    return kwargs
+
+def yaml_to_dict(tools: str) -> list[dict] | None:
+    try:
+        return yaml.safe_load(tools)
+    except YAMLError:
+        return None
+
+def extract_code(text: str) -> str:
+    pattern = r'```([^\n]*)\n(.*?)```'
+    matches = re.findall(pattern, text, re.DOTALL)
+    return matches[-1][1]
+
+# Append a conversation into history, while show it in a new markdown block
+def append_conversation(
+    conversation: Conversation,
+    history: list[Conversation],
+    placeholder: DeltaGenerator | None=None,
+) -> None:
+    history.append(conversation)
+    conversation.show(placeholder)
+
+def main(top_p: float, temperature: float, prompt_text: str, repetition_penalty: float):
+    manual_mode = st.toggle('Manual mode',
+        help='Define your tools in YAML format. You need to supply tool call results manually.'
+    )
+
+    if manual_mode:
+        with st.expander('Tools'):
+            tools = st.text_area(
+                'Define your tools in YAML format here:',
+                yaml.safe_dump([EXAMPLE_TOOL], sort_keys=False),
+                height=400,
+            )
+        tools = yaml_to_dict(tools)
+
+        if not tools:
+            st.error('YAML format error in tools definition')
+    else:
+        tools = get_tools()
+
+    if 'tool_history' not in st.session_state:
+        st.session_state.tool_history = []
+    if 'calling_tool' not in st.session_state:
+        st.session_state.calling_tool = False
+
+    history: list[Conversation] = st.session_state.tool_history
+
+    for conversation in history:
+        conversation.show()
+
+    if prompt_text:
+        prompt_text = prompt_text.strip()
+        role = st.session_state.calling_tool and Role.OBSERVATION or Role.USER
+        append_conversation(Conversation(role, prompt_text), history)
+        st.session_state.calling_tool = False
+
+        input_text = preprocess_text(
+            None,
+            tools,
+            history,
+        )
+        print("=== Input:")
+        print(input_text)
+        print("=== History:")
+        print(history)
+
+        placeholder = st.container()
+        message_placeholder = placeholder.chat_message(name="assistant", avatar="assistant")
+        markdown_placeholder = message_placeholder.empty()
+
+        for _ in range(5):
+            output_text = ''
+            for response in client.generate_stream(
+                system=None,
+                tools=tools,
+                history=history,
+                do_sample=True,
+                max_length=MAX_LENGTH,
+                temperature=temperature,
+                top_p=top_p,
+                stop_sequences=[str(r) for r in (Role.USER, Role.OBSERVATION)],
+                repetition_penalty=repetition_penalty,
+            ):
+                token = response.token
+                if response.token.special:
+                    print("=== Output:")
+                    print(output_text)
+
+                    match token.text.strip():
+                        case '<|user|>':
+                            append_conversation(Conversation(
+                                Role.ASSISTANT,
+                                postprocess_text(output_text),
+                            ), history, markdown_placeholder)
+                            return
+                        # Initiate tool call
+                        case '<|assistant|>':
+                            append_conversation(Conversation(
+                                Role.ASSISTANT,
+                                postprocess_text(output_text),
+                            ), history, markdown_placeholder)
+                            output_text = ''
+                            message_placeholder = placeholder.chat_message(name="tool", avatar="assistant")
+                            markdown_placeholder = message_placeholder.empty()
+                            continue
+                        case '<|observation|>':
+                            tool, *call_args_text = output_text.strip().split('\n')
+                            call_args_text = '\n'.join(call_args_text)
+                            
+                            append_conversation(Conversation(
+                                Role.TOOL,
+                                postprocess_text(output_text),
+                                tool,
+                            ), history, markdown_placeholder)
+                            message_placeholder = placeholder.chat_message(name="observation", avatar="user")
+                            markdown_placeholder = message_placeholder.empty()
+                            
+                            try:
+                                code = extract_code(call_args_text)
+                                args = eval(code, {'tool_call': tool_call}, {})
+                            except:
+                                st.error('Failed to parse tool call')
+                                return
+                            
+                            output_text = ''
+                            
+                            if manual_mode:
+                                st.info('Please provide tool call results below:')
+                                return
+                            else:
+                                with markdown_placeholder:
+                                    with st.spinner(f'Calling tool {tool}...'):
+                                        observation = dispatch_tool(tool, args)
+
+                                if len(observation) > TRUNCATE_LENGTH:
+                                    observation = observation[:TRUNCATE_LENGTH] + ' [TRUNCATED]'
+                                append_conversation(Conversation(
+                                    Role.OBSERVATION, observation
+                                ), history, markdown_placeholder)
+                                message_placeholder = placeholder.chat_message(name="assistant", avatar="assistant")
+                                markdown_placeholder = message_placeholder.empty()
+                                st.session_state.calling_tool = False
+                                break
+                        case _:
+                            st.error(f'Unexpected special token: {token.text.strip()}')
+                            return
+                output_text += response.token.text
+                markdown_placeholder.markdown(postprocess_text(output_text + '▌'))
+            else:
+                append_conversation(Conversation(
+                    Role.ASSISTANT,
+                    postprocess_text(output_text),
+                ), history, markdown_placeholder)
+                return
diff --git a/composite_demo/main.py b/composite_demo/main.py
new file mode 100644
index 0000000000000000000000000000000000000000..fbe13da07d77ff73ed5faa8cbfffd0726ecfa2e1
--- /dev/null
+++ b/composite_demo/main.py
@@ -0,0 +1,63 @@
+from enum import Enum
+import streamlit as st
+
+st.set_page_config(
+    page_title="ChatGLM3 Demo",
+    page_icon=":robot:",
+    layout='centered',
+    initial_sidebar_state='expanded',
+)
+
+import demo_chat, demo_ci, demo_tool
+
+DEFAULT_SYSTEM_PROMPT = '''
+You are ChatGLM3, a large language model trained by Zhipu.AI. Follow the user's instructions carefully. Respond using markdown.
+'''.strip()
+
+# Set the title of the demo
+st.title("ChatGLM3 Demo")
+
+# Add your custom text here, with smaller font size
+st.markdown("<sub>智谱AI 公开在线技术文档: https://lslfd0slxc.feishu.cn/wiki/WvQbwIJ9tiPAxGk8ywDck6yfnof </sub> \n\n <sub> 更多 ChatGLM3-6B 的使用方法请参考文档。</sub>", unsafe_allow_html=True)
+
+class Mode(str, Enum):
+    CHAT, TOOL, CI = '💬 Chat', '🛠️ Tool', '🧑‍💻 Code Interpreter'
+
+
+with st.sidebar:
+    top_p = st.slider(
+        'top_p', 0.0, 1.0, 0.8, step=0.01
+    )
+    temperature = st.slider(
+        'temperature', 0.0, 1.5, 0.95, step=0.01
+    )
+    repetition_penalty = st.slider(
+        'repetition_penalty', 0.0, 2.0, 1.2, step=0.01
+    )
+    system_prompt = st.text_area(
+        label="System Prompt (Only for chat mode)",
+        height=300,
+        value=DEFAULT_SYSTEM_PROMPT,
+    )
+
+prompt_text = st.chat_input(
+    'Chat with ChatGLM3!',
+    key='chat_input',
+)
+
+tab = st.radio(
+    'Mode',
+    [mode.value for mode in Mode],
+    horizontal=True,
+    label_visibility='hidden',
+)
+
+match tab:
+    case Mode.CHAT:
+        demo_chat.main(top_p, temperature, system_prompt, prompt_text, repetition_penalty)
+    case Mode.TOOL:
+        demo_tool.main(top_p, temperature, prompt_text, repetition_penalty)
+    case Mode.CI:
+        demo_ci.main(top_p, temperature, prompt_text, repetition_penalty)
+    case _:
+        st.error(f'Unexpected tab: {tab}')
diff --git a/composite_demo/requirements.txt b/composite_demo/requirements.txt
new file mode 100644
index 0000000000000000000000000000000000000000..27f603d0b554d4fa2df0c6e793ce1d1fead5e91a
--- /dev/null
+++ b/composite_demo/requirements.txt
@@ -0,0 +1,12 @@
+huggingface_hub
+ipykernel
+ipython
+jupyter_client
+pillow
+sentencepiece
+streamlit
+tokenizers
+torch
+transformers
+pyyaml
+requests
\ No newline at end of file
diff --git a/composite_demo/tool_registry.py b/composite_demo/tool_registry.py
new file mode 100644
index 0000000000000000000000000000000000000000..13d1478ff2f85d476c465320ccf246afa5281f99
--- /dev/null
+++ b/composite_demo/tool_registry.py
@@ -0,0 +1,109 @@
+import copy
+import inspect
+from pprint import pformat
+import traceback
+from types import GenericAlias
+from typing import get_origin, Annotated
+
+_TOOL_HOOKS = {}
+_TOOL_DESCRIPTIONS = {}
+
+def register_tool(func: callable):
+    tool_name = func.__name__
+    tool_description = inspect.getdoc(func).strip()
+    python_params = inspect.signature(func).parameters
+    tool_params = []
+    for name, param in python_params.items():
+        annotation = param.annotation
+        if annotation is inspect.Parameter.empty:
+            raise TypeError(f"Parameter `{name}` missing type annotation")
+        if get_origin(annotation) != Annotated:
+            raise TypeError(f"Annotation type for `{name}` must be typing.Annotated")
+        
+        typ, (description, required) = annotation.__origin__, annotation.__metadata__
+        typ: str = str(typ) if isinstance(typ, GenericAlias) else typ.__name__
+        if not isinstance(description, str):
+            raise TypeError(f"Description for `{name}` must be a string")
+        if not isinstance(required, bool):
+            raise TypeError(f"Required for `{name}` must be a bool")
+
+        tool_params.append({
+            "name": name,
+            "description": description,
+            "type": typ,
+            "required": required
+        })
+    tool_def = {
+        "name": tool_name,
+        "description": tool_description,
+        "params": tool_params
+    }
+
+    print("[registered tool] " + pformat(tool_def))
+    _TOOL_HOOKS[tool_name] = func
+    _TOOL_DESCRIPTIONS[tool_name] = tool_def
+
+    return func
+
+def dispatch_tool(tool_name: str, tool_params: dict) -> str:
+    if tool_name not in _TOOL_HOOKS:
+        return f"Tool `{tool_name}` not found. Please use a provided tool."
+    tool_call = _TOOL_HOOKS[tool_name]
+    try:
+        ret = tool_call(**tool_params)  
+    except:
+        ret = traceback.format_exc()
+    return str(ret)
+
+def get_tools() -> dict:
+    return copy.deepcopy(_TOOL_DESCRIPTIONS)
+
+# Tool Definitions
+
+@register_tool
+def random_number_generator(
+    seed: Annotated[int, 'The random seed used by the generator', True], 
+    range: Annotated[tuple[int, int], 'The range of the generated numbers', True],
+) -> int:
+    """
+    Generates a random number x, s.t. range[0] <= x < range[1]
+    """
+    if not isinstance(seed, int):
+        raise TypeError("Seed must be an integer")
+    if not isinstance(range, tuple):
+        raise TypeError("Range must be a tuple")
+    if not isinstance(range[0], int) or not isinstance(range[1], int):
+        raise TypeError("Range must be a tuple of integers")
+
+    import random
+    return random.Random(seed).randint(*range)
+
+@register_tool
+def get_weather(
+    city_name: Annotated[str, 'The name of the city to be queried', True],
+) -> str:
+    """
+    Get the current weather for `city_name`
+    """
+
+    if not isinstance(city_name, str):
+        raise TypeError("City name must be a string")
+
+    key_selection = {
+        "current_condition": ["temp_C", "FeelsLikeC", "humidity", "weatherDesc",  "observation_time"],
+    }
+    import requests
+    try:
+        resp = requests.get(f"https://wttr.in/{city_name}?format=j1")
+        resp.raise_for_status()
+        resp = resp.json()
+        ret = {k: {_v: resp[k][0][_v] for _v in v} for k, v in key_selection.items()}
+    except:
+        import traceback
+        ret = "Error encountered while fetching weather data!\n" + traceback.format_exc() 
+
+    return str(ret)
+
+if __name__ == "__main__":
+    print(dispatch_tool("get_weather", {"city_name": "beijing"}))
+    print(get_tools())
diff --git a/cookbook/accurate_prompt.ipynb b/cookbook/accurate_prompt.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..e69a92d2f279cb188e7bd51840132609b6abcac1
--- /dev/null
+++ b/cookbook/accurate_prompt.ipynb
@@ -0,0 +1,352 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "source": [
+    "# 使用精确的提示词完成符合格式输出的文本分类任务\n",
+    "\n",
+    "在本章节中，我们将带领开发者体验在 `文本分类` 任务中，在不进行任何模型微调的情况下，使用具有微小差异的提示词，并使用 `ChatGLM3-6B` 模型来完成符合格式输出的文本分类任务。\n",
+    "\n",
+    "我们使用 [新闻标题分类](https://github.com/fateleak/toutiao-multilevel-text-classfication-dataset) 任务来体验模型的表现。这是一个经典的文本分类任务，我们将使用 `新闻信息` 作为输入，模型需要预测出这个标题属于哪个类别。\n",
+    "\n",
+    "由于 `ChatGLM3-6B` 强大的能力，我们可以直接使用 `新闻标题` 作为输入，而不需要额外的信息，也不需要进行任何模型微调，就能完成这个任务。我们的目标是，让模型能够成功输出原始数据集中的15种类别中的一种作为分类结果，而且不能输入任何冗余的对话。\n",
+    "\n",
+    "在本章节中，用户将直观的对比两种不同细粒度的提示词下对模型分类造成的影响。\n",
+    "\n",
+    "## 硬件要求\n",
+    "本实践手册需要使用 FP16 精度的模型进行推理，因此，我们推荐使用至少 16GB 显存的 英伟达 GPU 来完成本实践手册。"
+   ],
+   "metadata": {
+    "collapsed": false
+   }
+  },
+  {
+   "cell_type": "markdown",
+   "source": [
+    "引入必要的库"
+   ],
+   "metadata": {
+    "collapsed": false
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "outputs": [],
+   "source": [
+    "from transformers import AutoModel, AutoTokenizer\n",
+    "import torch"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-11-23T19:23:03.002351Z",
+     "end_time": "2023-11-23T19:23:03.016701Z"
+    }
+   }
+  },
+  {
+   "cell_type": "markdown",
+   "source": [
+    "设置好对应的参数，以保证模型推理的公平性。"
+   ],
+   "metadata": {
+    "collapsed": false
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "outputs": [],
+   "source": [
+    "max_new_tokens = 1024\n",
+    "temperature = 0.1\n",
+    "top_p = 0.9\n",
+    "device = \"cuda\"\n",
+    "model_path_chat = \"/Models/chatglm3-6b\""
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-11-23T19:23:03.004850Z",
+     "end_time": "2023-11-23T19:23:03.047154Z"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "Loading checkpoint shards:   0%|          | 0/7 [00:00<?, ?it/s]",
+      "application/vnd.jupyter.widget-view+json": {
+       "version_major": 2,
+       "version_minor": 0,
+       "model_id": "d6468f7889994e638ac754fde95b6e58"
+      }
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "tokenizer_path_chat = model_path_chat\n",
+    "tokenizer = AutoTokenizer.from_pretrained(tokenizer_path_chat, trust_remote_code=True, encode_special_tokens=True)\n",
+    "model = AutoModel.from_pretrained(model_path_chat, load_in_8bit=False, trust_remote_code=True).to(device)"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-11-23T19:23:03.047154Z",
+     "end_time": "2023-11-23T19:23:13.086148Z"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "outputs": [],
+   "source": [
+    "def answer(prompt):\n",
+    "    inputs = tokenizer(prompt, return_tensors=\"pt\").to(device)\n",
+    "    response = model.generate(input_ids=inputs[\"input_ids\"], max_new_tokens=max_new_tokens, history=None,\n",
+    "                              temperature=temperature,\n",
+    "                              top_p=top_p, do_sample=True)\n",
+    "    response = response[0, inputs[\"input_ids\"].shape[-1]:]\n",
+    "    answer = tokenizer.decode(response, skip_special_tokens=True)\n",
+    "    return answer"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-11-23T19:23:13.086148Z",
+     "end_time": "2023-11-23T19:23:13.086148Z"
+    }
+   }
+  },
+  {
+   "cell_type": "markdown",
+   "source": [
+    "在本样例中，模型应该输出的标准答案应该为:\n",
+    "'news_sports'"
+   ],
+   "metadata": {
+    "collapsed": false
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "outputs": [],
+   "source": [
+    "PROMPT1 = \"\"\"\n",
+    "<|system|>\n",
+    "你是一个专业的新闻专家，请根据我提供的新闻信息，包括新闻标题，新闻关键词等信息，你需要在这些类中选择其中一个，它们分别是:\n",
+    "news_story\n",
+    "news_culture\n",
+    "news_sports\n",
+    "news_finance\n",
+    "news_house\n",
+    "news_car\n",
+    "news_edu\n",
+    "news_tech\n",
+    "news_military\n",
+    "news_travel\n",
+    "news_world\n",
+    "stock\n",
+    "news_agriculture\n",
+    "news_game\n",
+    "这是我的信息:\n",
+    "<|user|>\n",
+    "新闻标题: 女乒今天排兵布阵不合理，丁宁昨天刚打硬仗，今天应该打第一单打，你认为呢？\n",
+    "新闻关键词: 无\n",
+    "<|assistant|>\n",
+    "\"\"\""
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-11-23T19:23:13.086148Z",
+     "end_time": "2023-11-23T19:23:13.095864Z"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "' 新闻类型：体育新闻\\n  新闻关键词：女乒，排兵布阵，丁宁，硬仗，第一单打'"
+     },
+     "execution_count": 24,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "answer(PROMPT1)"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-11-23T19:23:13.095864Z",
+     "end_time": "2023-11-23T19:23:14.034154Z"
+    }
+   }
+  },
+  {
+   "cell_type": "markdown",
+   "source": [
+    "很明显，使用这个提示词没有完成要求。我们更换一个提示词，经过优化后，看是否能达标。\n",
+    "\n",
+    "我们为提示词中设定了更多的限定词汇，并用Markdown语法规范化了输出的格式。修改后的提示词如下："
+   ],
+   "metadata": {
+    "collapsed": false
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "outputs": [],
+   "source": [
+    "PROMPT2 = \"\"\"\n",
+    "<|system|>\n",
+    "请根据我提供的新闻信息，格式为\n",
+    "```\n",
+    "新闻标题: xxx\n",
+    "新闻关键词: xxx\n",
+    "```\n",
+    "你要对每一行新闻类别进行分类并告诉我结果，不要返回其他信息和多于的文字，这些类别是:\n",
+    "news_story\n",
+    "news_culture\n",
+    "news_sports\n",
+    "news_finance\n",
+    "news_house\n",
+    "news_car\n",
+    "news_edu\n",
+    "news_tech\n",
+    "news_military\n",
+    "news_travel\n",
+    "news_world\n",
+    "stock\n",
+    "news_agriculture\n",
+    "news_game\n",
+    "我将为你提供一些新闻标题和关键词，你需要根据这些信息，对这些新闻进行分类，每条一行，格式为\n",
+    "```\n",
+    "类别中的其中一个\n",
+    "```\n",
+    "不要返回其他内容，例如新闻标题，新闻关键词等等，只需要返回分类结果即可\n",
+    "<|user|>\n",
+    "新闻标题: 女乒今天排兵布阵不合理，丁宁昨天刚打硬仗，今天应该打第一单打，你认为呢？\n",
+    "新闻关键词: 无\n",
+    "<|assistant|>\n",
+    "\"\"\""
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-11-23T19:23:14.044595Z",
+     "end_time": "2023-11-23T19:23:14.044595Z"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "'news_sports'"
+     },
+     "execution_count": 26,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "answer(PROMPT2)"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-11-23T19:23:14.044595Z",
+     "end_time": "2023-11-23T19:23:14.217801Z"
+    }
+   }
+  },
+  {
+   "cell_type": "markdown",
+   "source": [
+    "这一次，模型成功给出了理想的答案。我们结束实操训练，删除模型并释放显存。"
+   ],
+   "metadata": {
+    "collapsed": false
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "outputs": [],
+   "source": [
+    "del model\n",
+    "torch.cuda.empty_cache()"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-11-23T19:23:14.217801Z",
+     "end_time": "2023-11-23T19:23:14.373137Z"
+    }
+   }
+  },
+  {
+   "cell_type": "markdown",
+   "source": [
+    "## 总结\n",
+    "在本实践手册中，我们让开发者体验了使用不同提示词下，`ChatGLM3-6B` 模型在 `新闻标题分类` 任务中的表现。\n",
+    "1. 模型在两次提示中都能正确的分类，验证了模型底座的能力。\n",
+    "2. 在使用第二个提示词时，模型的输出符合格式要求，而且没有冗余的对话。符合我们的要求。\n",
+    "3. 第二个提示词使用了一定的表示限定的词汇和更准确的任务要求，包括规定了返回格式，返回的行书等的。\n",
+    "\n",
+    "因此，通过上述内容，我们可以发现：\n",
+    "对于有格式要求的任务，可以使用更具有格式化的提示词来完成任务。同时，使用更低的 `temperature` 和更高的 `top_p` 可以提高模型的输出质量。\n",
+    "\n",
+    "模型在训练中大量使用 Markdown格式。因此，用 ``` 符号来规范提示词将能提升模型输出格式化数据的准确率。"
+   ],
+   "metadata": {
+    "collapsed": false
+   }
+  },
+  {
+   "cell_type": "markdown",
+   "source": [],
+   "metadata": {
+    "collapsed": false
+   }
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 2
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython2",
+   "version": "2.7.6"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 0
+}
diff --git a/cookbook/data/toutiao_cat_data_example.txt b/cookbook/data/toutiao_cat_data_example.txt
new file mode 100644
index 0000000000000000000000000000000000000000..5be76dbbe1ec3d7560da6f1b45167dd8bc1e366e
--- /dev/null
+++ b/cookbook/data/toutiao_cat_data_example.txt
@@ -0,0 +1,4999 @@
+6551700932705387022_!_101_!_news_culture_!_京城最值得你来场文化之旅的博物馆_!_保利集团,马未都,中国科学技术馆,博物馆,新中国
+6552368441838272771_!_101_!_news_culture_!_发酵床的垫料种类有哪些？哪种更好？_!_
+6552407965343678723_!_101_!_news_culture_!_上联：黄山黄河黄皮肤黄土高原。怎么对下联？_!_
+6552332417753940238_!_101_!_news_culture_!_林徽因什么理由拒绝了徐志摩而选择梁思成为终身伴侣？_!_
+6552475601595269390_!_101_!_news_culture_!_黄杨木是什么树？_!_
+6552387648126714125_!_101_!_news_culture_!_上联：草根登上星光道，怎么对下联？_!_
+6552271725814350087_!_101_!_news_culture_!_什么是超写实绘画？_!_
+6552452982015787268_!_101_!_news_culture_!_松涛听雨莺婉转，下联？_!_
+6552400379030536455_!_101_!_news_culture_!_上联：老子骑牛读书，下联怎么对？_!_
+6552339283632455939_!_101_!_news_culture_!_上联：山水醉人何须酒。如何对下联？_!_
+6552387367334838792_!_101_!_news_culture_!_国画山水，如何读懂山水画_!_林风眠,黄海归来步步云,秋山图,计白当黑,山水画,江山万里图,张大千,巫峡清秋图,活眼,山雨欲来图
+6552314684383429128_!_101_!_news_culture_!_一元硬币是这种，现在价值24000元，找找看！_!_牡丹,收藏价值
+6552128476109865229_!_101_!_news_culture_!_有哪些让人感动的语句呢？_!_
+6552447749072093453_!_101_!_news_culture_!_上联，绿竹引清风，如何对下联？_!_
+6552301380562846215_!_101_!_news_culture_!_赵宁安作品欣赏_!_叶浅予,田世光,李苦禅,花鸟画,中央美术学院
+6552263394420850951_!_101_!_news_culture_!_夕阳无语燕归愁，如何接下句？_!_
+6552290314294395139_!_101_!_news_culture_!_上联：山水醉人何须酒。如何对下联？_!_
+6552462208314376462_!_101_!_news_culture_!_上联：上班为下班，如何对下联？_!_
+6552311866947797262_!_101_!_news_culture_!_下联:夕陽西下已黄昏。上联是什麽？_!_
+6552466638304707079_!_101_!_news_culture_!_初夏方好，小荷初露，一起来读那些美不胜收的咏荷经典诗词_!_荷花,西湖,金粟词话,采莲女,念奴娇·赤壁怀古,林逋,荷叶
+6552431613437805063_!_102_!_news_entertainment_!_谢娜为李浩菲澄清网络谣言，之后她的两个行为给自己加分_!_佟丽娅,网络谣言,快乐大本营,李浩菲,谢娜,观众们
+6552320560913711629_!_102_!_news_entertainment_!_谢娜曾为他与主办方撕破脸！谢娜复出快本！他第一个到场支持！_!_汪涵,火星情报局,杨迪,主办方,谢娜,刘维
+6552390546051039747_!_102_!_news_entertainment_!_中国网红竟红到美国？不多说了，连小编都心动了_!_飞纱,新娘,脱口秀,中国网,婚礼
+6552150358678831624_!_102_!_news_entertainment_!_赵丽颖很久没有登上微博热搜了，但你们别急，她只是在憋大招而已_!_陆贞传奇,大红大紫,楚乔传,微博热搜,赵丽颖,花千骨,迪丽热巴,Angelababy
+6552408585177924099_!_102_!_news_entertainment_!_因戴一个眼镜更改变气质的6大娱乐圈明星，你最喜欢哪一个_!_戴上眼镜,刘德华,张翰,远大前程,杜志国,刘亦菲
+6552147830184608263_!_102_!_news_entertainment_!_后来的我们，抢先看_!_电影院,前任3,刘若英,张一白,田壮壮
+6552472345548685837_!_102_!_news_entertainment_!_超级英雄演员颜值身材排名，钢铁侠进不了前5，第一名很意外_!_金刚狼3,休·杰克曼,神奇女侠,绯红女巫,超人,金刚狼
+6552385284682547716_!_102_!_news_entertainment_!_《无限歌谣季》热播 张绍刚毛不易组合似“父子”_!_张绍刚,新组合,腾讯视频,无限歌谣季,毛不易,父子
+6552364464715334151_!_102_!_news_entertainment_!_张靓颖透露右耳已经间歇性失聪10年，这些年她都是怎么过来的啊_!_中岛美嘉,滨崎步,张靓颖,演唱会,林子祥
+6552310157706002702_!_102_!_news_entertainment_!_成龙改口决定不裸捐了，20亿财产给儿子一半，你怎么看？_!_
+6552286735408038403_!_102_!_news_entertainment_!_五大“出轨“女明星，最后一个你们肯定不知道！_!_王全安,张柏芝,张雨绮,吴卓林,谢霆锋
+6552269871697101315_!_102_!_news_entertainment_!_认真搞笑的男人最帅！大张伟不张嘴就是美男子之最！蜜汁好看~_!_天天向上,毒鸡汤,大张伟,美男子,综艺,我去上学了,百变大咖秀
+6552418723179790856_!_102_!_news_entertainment_!_谢娜三喜临门，何炅送祝福，吴昕送祝福，只有沈梦辰不一样_!_杜海涛,谢娜,何炅,沈梦辰,吴昕,快本
+6552283654494617859_!_102_!_news_entertainment_!_如何评价赵丽颖？_!_
+6552453686398812423_!_102_!_news_entertainment_!_有哪些偏冷门的歌曲推荐？_!_
+6552383324696871427_!_102_!_news_entertainment_!_“整容狂人”的审美，恕欣赏不来_!_高富帅,阿尔维斯,颜值,吴彦祖,罗德里戈
+6552390851157295629_!_102_!_news_entertainment_!_杨幂景甜徐冬冬唐嫣 不好好穿衣却美的有趣又撩人_!_杨幂,徐冬冬,背带裙,大唐荣耀,唐嫣,景甜
+6552445894711575043_!_102_!_news_entertainment_!_吴卓林：你父母固然有责任，但最大的责任还是在于你自己！_!_孤儿院,吴卓林,小龙女,加拿大
+6552351831668818435_!_102_!_news_entertainment_!_《天乩战之白蛇传说》赵雅芝和杨紫饰演母女会擦出什么样的火花？_!_欢乐颂,西王母,老戏骨,赵雅芝,杨紫,刘嘉玲,新白娘子传奇
+6552474018920792580_!_102_!_news_entertainment_!_他是最帅反派专业户，演《古惑仔》大火，今病魔缠身可怜无人识！_!_浩南哥,黄秋生,龙虎风云,东陵大盗,爱恋狂潮,李子雄,锣鼓巷,古惑仔,学校风云,郑伊健,张耀扬,监狱风云,无间道2
+6552309039697494532_!_103_!_news_sports_!_亚洲杯夺冠赔率：日本、伊朗领衔 中国竟与泰国并列_!_土库曼斯坦,乌兹别克斯坦,亚洲杯,赔率,小组赛
+6552477789642031623_!_103_!_news_sports_!_9轮4球本土射手仅次武磊 黄紫昌要抢最强U23头衔_!_黄紫昌,武磊,卡佩罗,惠家康,韦世豪
+6552495859798376712_!_103_!_news_sports_!_如果今年勇士夺冠，下赛季詹姆斯何去何从？_!_
+6552202204621570574_!_103_!_news_sports_!_超级替补！科斯塔本赛季替补出场贡献7次助攻_!_替补出场,助攻,科斯塔,赫迪拉,博洛尼亚
+6552409639911162381_!_103_!_news_sports_!_骑士6天里发生了啥？从首轮抢七到次轮3-0猛龙_!_骑士,半决赛,猛龙,德罗赞,乐福
+6552470008188895491_!_103_!_news_sports_!_如果朗多进入转会市场，哪些球队适合他？_!_
+6552316209847599367_!_103_!_news_sports_!_詹姆斯G3决杀，你怎么看？_!_
+6552295807171691022_!_103_!_news_sports_!_大魔王带头唱歌！火箭这像是打季后赛？爵士神帅这话已提前投降了_!_PJ-塔克,斯奈德,大魔王,火箭,爵士
+6552452056043487748_!_103_!_news_sports_!_马夏尔要去切尔西？可以商量，不过穆里尼奥的要价是4000万加威廉_!_威廉,曼联,穆里尼奥,布莱顿,马夏尔
+6552287279061139972_!_103_!_news_sports_!_利希施泰纳宣布赛季结束后离队：我需要新的挑战_!_尤文,博洛尼亚,施泰纳,利希施泰纳,拉齐奥
+6552325997813825805_!_103_!_news_sports_!_怎么样看待大连一方在中超联赛第九轮取得的赛季首胜？_!_
+6552468503452975630_!_103_!_news_sports_!_科勒·卡戴珊与男友TT共进午餐，曾在他怀孕期间偷腥的渣男被原_!_NBA,科勒,卡戴珊,四季酒店,克利夫兰骑士队
+6552312104492204301_!_103_!_news_sports_!_作为央视体育体育频道，CCTV5一到周末就直播马拉松你怎么看？_!_
+6552332687678374151_!_103_!_news_sports_!_如果2018骑士夺冠，詹姆斯这个冠军的含金量有多大？_!_
+6552466727995703815_!_103_!_news_sports_!_昔日中超金靴半场独造6球虐爆辽足 华夏送走他后悔吗？_!_阿洛,阿洛伊西奥,华夏幸福,埃尔纳内斯,穆里奇
+6552416247298916612_!_103_!_news_sports_!_NBA历史排名前十都有谁？_!_
+6552354211424633352_!_103_!_news_sports_!_你希望利物浦赢得欧冠吗？巴萨主帅巴尔韦德的回答耐人寻味_!_双冠王,欧冠赛场,贝里索,巴尔韦德,欧冠冠军
+6552314862163198471_!_103_!_news_sports_!_绝杀！詹姆斯38+7再创传奇一刻，两护法创另类神迹更功不可没_!_猛龙,詹姆斯,伊巴卡,德罗赞,科沃尔
+6552304431164031502_!_103_!_news_sports_!_再现绝杀！今天的老詹怎么吹_!_德罗赞,詹姆斯
+6552404265724281357_!_103_!_news_sports_!_拜仁3比1逆转科隆，J罗现世界级做饼_!_J罗,科隆,拜仁,假动作,托利索
+6552375446955098627_!_104_!_news_finance_!_法治聚焦｜多部门联合发文规范民间借贷，严厉打击暴力催收_!_金融,民间借贷,法治
+6552249357389791758_!_104_!_news_finance_!_中国最独特的公司，温氏究竟厉害在哪里？_!_互联网企业,养猪,分布式,家庭农场,新兴县,温氏集团,温北英,可口可乐,技术创新,周其仁,农业
+6552392199802192391_!_104_!_news_finance_!_投P2P本金加12%利息全打水漂 后来我开始研究贷款_!_信用卡,股票,银行,P2P,利息,余额宝
+6552423505156112899_!_104_!_news_finance_!_5月31号前，企业需完成汇算清缴工作，会计注意这5个细节问题_!_汇算清缴,纳税人,残疾人,汇总纳税企业,企业所得税税款
+6552393713924964872_!_104_!_news_finance_!_下周一（5.7日）手上持有这些股的要小心_!_股权解禁,新五丰,鲁阳节能,解禁,大小非
+6552284551496859908_!_104_!_news_finance_!_5万元资金，该做什么行业？_!_
+6552465616924574222_!_104_!_news_finance_!_信用卡账单免息期与最长50天免息还款技巧大全！_!_还款日,账单日,还款,信用卡,免息期
+6552379453605937422_!_104_!_news_finance_!_美增加关税，为何人民币不降反升？_!_
+6552381513936142862_!_104_!_news_finance_!_冯天睿：5.6千三大关何以告破 黄金原油下周一操作建议_!_心静如水,大阳线,MACD,操作,黄金,原油
+6552128165496488456_!_104_!_news_finance_!_报名｜界面资本论坛热点前瞻：各界共议“创造可持续价值”_!_海兰信,安永,中国企业,上海电气,一带一路,交通银行,创造可持续价值,上市,华建集团,阿里巴巴,国药控股
+6552367589337596174_!_104_!_news_finance_!_2017年的巴菲特股东大会有哪些看点？_!_
+6552171081761817091_!_104_!_news_finance_!_巴菲特：美国人生活会越来越好 比洛克菲勒都要过得好_!_美国,金融界,巴菲特,民主党,洛克菲勒
+6552135862929326600_!_104_!_news_finance_!_践行“军民融合”，助力“一带一路”_!_中国电子信息产业集团,电子对抗,美亚柏科,雷科防务,华力创通,中国电子,旋极信息,军民融合,民营企业,中国电子科技集团公司
+6552348488141636099_!_104_!_news_finance_!_重磅：“5050”计划打造全国人才生态最优区 “国际滨”蓄势待发_!_5050计划,高新区,创新创业,阿里巴巴,滨江,杭州市
+6552493681381736963_!_104_!_news_finance_!_不仅首季业绩持续高增长，这些股中期业绩仍可期_!_季度业绩,业绩预告,华帝股份,上证指数,净利润增幅,业绩增长,高增长,有色金属
+6552479093923774734_!_104_!_news_finance_!_目前有35万存款，怎样在短期之内翻倍？_!_
+6552387539792036360_!_104_!_news_finance_!_中兴通讯：关于重大事项进展公告_!_中兴通讯,BIS,中兴通讯股份有限公司
+6552474712801280520_!_104_!_news_finance_!_好好把握比特币的调整 每次都是机会_!_巴菲特,虚拟货币,伯克希尔哈撒韦公司,伯克希尔哈撒韦,比特币
+6552356541591192067_!_104_!_news_finance_!_寻找地产界“最强大脑”——中国房企2018年度CFO评选启幕_!_中国房地产,最强大脑,超级CFO,房地产行业,房地产
+6552451858437243395_!_104_!_news_finance_!_独角兽药明康德周二上市，但一季报净利已现业绩下滑_!_巨人网络,药明康德,分众传媒,招股书,独角兽企业
+6552303945463628301_!_106_!_news_house_!_杭州哪里还有“1万+”的新房？优先选择哪一个？_!_摇号,未来科技城,阳光城,小高层,酒店式公寓
+6552357571120857607_!_106_!_news_house_!_三明贵溪洋拟建服务型公寓，共300多套！附市区5月份二手房价_!_徐碧街道,宏图花园,城乡规划局,村民委员会,二手房
+6552472831328780803_!_106_!_news_house_!_互联侃谈：霍官泰和他的楼花，房地产的行业，到底有多深？_!_期房,何鸿燊,分层产权制度,霍官泰,楼花
+6552418506741121544_!_106_!_news_house_!_山东最新17市的房价出炉了！快看，你家房价的走向！是真高！_!_房价,山东省第一季度薪酬表,二手房价格行政区二手房,山东,二手房
+6552413070562427144_!_106_!_news_house_!_物业公司的服务水准非常烂，该怎么解决？_!_
+6552300361216950531_!_106_!_news_house_!_实行房产税和遗产税之后小产权房怎么办？_!_
+6552327554336817678_!_106_!_news_house_!_成都摇号选房·天立世纪华府——彭州清水均价5476，不是投资处_!_都江堰,彭州,楼盘,天立世纪华府,㎡三室两厅两卫户型
+6552296976585589251_!_106_!_news_house_!_「头条」2018-金色里程，最新详情解析！马鞍山当涂住宅！_!_当涂,城市发展规划,金色里程,马鞍山,当涂县,马鞍山市
+6552301698050687501_!_106_!_news_house_!_瞭望刊文：房地产税改革渐行渐近 还需突破四个阻力区_!_税改革,土地财政,产权房,胡光辉,瞭望,不动产,房地产税,税费
+6552494403947069704_!_106_!_news_house_!_四川巴中有哪些好的装修队？_!_
+6551907587439198723_!_106_!_news_house_!_房屋被征收方强拆，已过起诉时效，还有什么救济方法？_!_行政行为,诉讼时效,行政机关
+6552272429509509389_!_106_!_news_house_!_惠州惠阳13000一平的房子值得买吗？_!_
+6552403967991611662_!_106_!_news_house_!_铜梁的房价是多少？_!_
+6552316609770291726_!_106_!_news_house_!_售楼闺蜜告诉我，这几种户型的房子万万别买，后悔我家买错了_!_房价,户型,售楼,买房
+6552397419261198606_!_106_!_news_house_!_未来房价会跌到什么程度？_!_
+6552387849344254222_!_106_!_news_house_!_未来房价会跌到什么程度？_!_
+6552295326462509575_!_106_!_news_house_!_2018年的房价回是什么趋势？_!_房地产泡沫,海南岛,房价,大洗牌,房地产开发商
+6552327314498126083_!_106_!_news_house_!_五月份是买房的好时间吗？_!_
+6552114549741322766_!_106_!_news_house_!_火爆的楼市现状，刚需最后的上车机会？_!_保利新武昌,首套房,购房者,购房合同,利率上浮
+6552236298344595982_!_106_!_news_house_!_青岛这个区牛了！要建奥特莱斯、银座……还通地铁！_!_世茂,银座,高新区,奥特莱斯,青岛高新区,项目投资协议
+6552322499793649924_!_107_!_news_car_!_将SUV和轿车融合在一起的新款沃尔沃，值得买吗？_!_
+6552468141958496781_!_107_!_news_car_!_又一款国产中大型7座SUV将于后天正式上市，你喜欢吗？_!_众泰T800,LED光源,T700,众泰,T800,众泰T700,众泰汽车
+6552475198212276750_!_107_!_news_car_!_名爵ZS纯电版车型曝光，电子旋钮换挡，2018下半年上市！_!_SUV,名爵ZS,名爵品牌,名爵
+6552134625018249735_!_107_!_news_car_!_重机云集的醉美江南机车节上谁“最美”？胡军李亚鹏告诉你答案_!_李亚鹏,印第安,Bobber,北极星,胡军,车友
+6552100339200295438_!_107_!_news_car_!_地表最强1.2T自主紧凑SUV即将上市，动力甩合资几条街_!_发动机,海马S5,海马,海马汽车,第二代,福美来,速双离合,双离合变速箱,DVVT
+6552410907182366990_!_107_!_news_car_!_裸车价10万以内家用轿车有什么推荐？_!_
+6552363085091307780_!_107_!_news_car_!_荣威MARVEL X首发如何做到“三过硬五第一”？_!_
+6552474475613389063_!_107_!_news_car_!_同为2.0t的VV5和RX5该怎么选？哪款性价比高？_!_
+6552247840540721671_!_107_!_news_car_!_越来越多夫妻自驾游不去酒店，义乌造“房车神器”，真是大开眼界_!_充气床,车睡神器,防眩镜,烟灰缸,房车
+6552489746281529613_!_107_!_news_car_!_五菱宏光可以换发动机吗？噪音有点大，想换个大众的发动机，可行吗？_!_
+6551548654044316163_!_107_!_news_car_!_曝光台：违反禁令标志_!_道路交通安全法,海南省,禁令标志
+6552397614560576014_!_107_!_news_car_!_众泰这次又火了，全新车标比“玛莎拉蒂”还帅气！_!_保时捷Macan,Macan,T700,众泰T600,车标,奥迪Q3,保时捷,众泰,众泰汽车,玛莎拉蒂
+6552406371462021635_!_107_!_news_car_!_「新车」国产奔驰A级三厢正式亮相 豪华紧凑型轿车的新标杆_!_人机交互,涡轮增压,A-Class,速双离合,级三厢,Sedan
+6552332997431919111_!_107_!_news_car_!_在华销量破10万，限量版跑车售罄，CX-8成热点，马自达信心大增_!_长安马自达,长安,马自达,一汽,MX-5,CX-8,一汽马自达,创驰蓝天
+6552253258084123150_!_107_!_news_car_!_夏季暴晒很伤车！精明领导这样做，汽车寿命多延续5年，真凉爽_!_防眩镜,反射镜,玻璃水,HUD,护目镜
+6552441452339659267_!_107_!_news_car_!_5月10日起，德州开发区这个路口开始抓拍“不礼让斑马线”！_!_道路交通,人行横道,机动车,中华人民共和国道路交通安全法,道路交通安全法实施条例,道路交通安全法,交通安全法
+6552382749393551885_!_107_!_news_car_!_汉腾汽车首款MPV，想做车市黑马，先问问宝骏730同意不_!_MPV,中控台,GM8,R18,汉腾旗下
+6552416756030243080_!_107_!_news_car_!_十万元能上路的汽车，买什么比较好？_!_
+6552296183421731336_!_107_!_news_car_!_李嘉诚这次火了，投资51亿，新车颜值不输特斯拉，董明珠都羡慕！_!_雷克萨斯,李嘉诚,新车,董明珠,保险杠,特斯拉
+6552307079355302408_!_107_!_news_car_!_那天，我亲自送老公去开房_!_异地恋,陆风,打工妹,陈佑林,分手信
+6552294650537837070_!_108_!_news_edu_!_奥数天才当教师，人生换轨，何来坠落_!_伤仲永,人物,军事理论,马克思,IMO,奥数天才坠落之后,北京大学,付云皓
+6552372463005598212_!_108_!_news_edu_!_儿子的同学借给他几本书，说是作文想得“奖”，熟记下来准没错_!_彩页印刷,阅读障碍,写作方法,全套10册
+6552290807167058184_!_108_!_news_edu_!_杨振宁说：把中学生送到美国读书非常危险，你怎么评价？_!_
+6552328105527083268_!_108_!_news_edu_!_猪伪狂犬苗的免疫方案怎么做？_!_
+6552401901990707469_!_108_!_news_edu_!_孩子英语不好，时态分不清，该怎么办？_!_
+6552313119366971651_!_108_!_news_edu_!_没有教不会的学生，只有不会教的老师，这说法正确吗？_!_
+6552282576759816455_!_108_!_news_edu_!_合肥有哪些计算机培训学习机构？_!_
+6552420799335104775_!_108_!_news_edu_!_基金从业资格证书的含金量怎么样？_!_
+6552405148759490823_!_108_!_news_edu_!_找工作要学历,学历等于能力吗？_!_
+6552377382408290830_!_108_!_news_edu_!_北大校长口误念错发音 发表致歉信公开道歉_!_矛盾论,林建华,大学校长,实践论,北京大学,对立统一,老三篇
+6551546790531826189_!_108_!_news_edu_!_英汉句子：这个世界，没有偶然_!_汉语,英语原
+6552449759905317128_!_108_!_news_edu_!_西南大学的师范专业在师范类排在什么位置？_!_
+6552484893996089870_!_108_!_news_edu_!_奥运冠军领跑，省实南海5000家长学生共赴亲子跑_!_李珊珊,记者陈芳庭,跑活动,广东实验中学,平衡木
+6552281009172250884_!_108_!_news_edu_!_你见过最努力的人现在都变成什么样子了？_!_
+6552365402876281091_!_108_!_news_edu_!_高三学生如何合理安排暑假？_!_
+6552293050754793992_!_108_!_news_edu_!_江西高考重大发布！关于录取和志愿填报，看这一篇就够了！_!_高考,体育类,三校生,语文,提前批
+6552394766321975816_!_108_!_news_edu_!_没有爱，就没有教育！江门开平一幼儿园收到家长亲笔写的“情书”_!_美丽中国,幼儿园,牛津,江门,Nini,感谢信
+6552355723748049166_!_108_!_news_edu_!_肥沃的近义词是什么？_!_
+6552471417135301133_!_108_!_news_edu_!_特警进重庆川外当老师，手把手教“女子防身术”_!_四川外国语大学,薛雪,重庆,平安课堂,特警总队
+6552309349262295560_!_108_!_news_edu_!_就业对接直入校园 广州大学城创新招聘会吸引过千人参加_!_广州大学,人事行政,大学城,大学生,招聘会
+6552443882590699783_!_109_!_news_tech_!_苹果公司现金流近三千亿美元，会有收购阿里巴巴的可能性吗？_!_
+6552384446178263560_!_109_!_news_tech_!_15 分钟参透比特币和区块链_!_随机数,区块链,链条,矿工,挖矿,广播,交易记录,去中心化,电子货币,账本
+6552485591915692292_!_109_!_news_tech_!_微信支付是否已远远超过支付宝？_!_
+6552280614005899523_!_109_!_news_tech_!_如果腾讯不再免费，QQ和微信月租50元，中国还会有多少人继续用？_!_
+6552296864044024077_!_109_!_news_tech_!_如果所有银行不让支付宝绑定银行卡了，支付宝会怎样？_!_
+6552320508447162628_!_109_!_news_tech_!_马化腾回应腾讯没有梦想，他回应说的话符合他一直做的吗？_!_
+6552482135226188039_!_109_!_news_tech_!_京东市值3个月蒸发1200亿，为何屡遭机构唱空？_!_
+6552319168148930823_!_109_!_news_tech_!_未来取代app通信的方向在哪里？_!_
+6552496405670265348_!_109_!_news_tech_!_阿里巴巴准备扩张东南亚市场_!_东南亚,张丹丹,马云,阿里巴巴,小型企业
+6552130347214045710_!_109_!_news_tech_!_Fullestop调研VR技术在不同行业都存在影响力_!_网页设计,VR,虚拟现实,电子游戏,纽约市
+6552281358536802820_!_109_!_news_tech_!_网友数博会：大数据与实体经济融合发展，贵州焕发新动力_!_大数据,贵州,数字经济,实体,老干妈
+6552448661727478030_!_109_!_news_tech_!_手机天气软件哪个最好？_!_
+6552268643076407812_!_109_!_news_tech_!_净网2018，打击网络黑色产业链_!_净网2018,李旭阳,网络犯罪,人工智能,打击网络
+6552273121020215556_!_109_!_news_tech_!_你如何看待央视盛赞支付宝？_!_
+6552312594357551367_!_109_!_news_tech_!_玉米在播种前，需要二次包衣吗，为什么呢？_!_
+6552370092556943629_!_109_!_news_tech_!_偶尔玩游戏，小米mix2续航怎么样？_!_
+6552425259633475843_!_109_!_news_tech_!_北京西二旗月薪5万的程序员真实生活状态什么样子的？_!_
+6552357649327849988_!_109_!_news_tech_!_深圳地铁“扫码过闸”下周二正式上线_!_深圳地铁,钛媒体
+6552360988279046404_!_109_!_news_tech_!_哪些行为在产品开发中是耍流氓？_!_
+6552342158215479565_!_109_!_news_tech_!_联想被踢出恒生指数，是什么导致了联想现在的这种境地？_!_
+6552400743796572429_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552172312798429444_!_110_!_news_military_!_如果在距离地球500公里的太空引爆一颗五千万弹量的氢弹，会怎么样？_!_
+6552452526594064643_!_110_!_news_military_!_日本的93式氧气鱼雷威力为什么那么大？_!_
+6552312289586839816_!_110_!_news_military_!_军演时来敌人了怎么办？_!_
+6552387440152150536_!_110_!_news_military_!_美出台国防授权法案应对中俄竞争，国防预算超7000亿美元_!_理查森,北大西洋,俄罗斯,网络战,司令部
+6552347705409011972_!_110_!_news_military_!_越南拥有110余万军队，为何越南军队最擅长布置竹签阵？_!_
+6552290168697520654_!_110_!_news_military_!_美军终于怕了：彩虹无人机已实战干掉300多目标_!_彩虹5,导弹,攻击机,最大起飞重量,无人机
+6551731822361838088_!_110_!_news_military_!_雄风不再：俄罗斯图-22M轰炸机群已现旧态_!_沃罗涅日,俄罗斯,俄军比武,战略轰炸机,轰炸机
+6552246212412572173_!_110_!_news_military_!_美航母刚在叙利亚空袭，俄军就果断亮剑：美军这下危险了_!_俄军,俄罗斯,叙利亚,图-160,导弹,战机
+6552306716527034627_!_110_!_news_military_!_东风21D的战略威慑究竟有多大？_!_
+6552275277039272200_!_110_!_news_military_!_你怎么看待以色列总理出席俄罗斯5月9日的阅兵并会晤普京？_!_
+6552131033687392519_!_110_!_news_military_!_如果美国退出联合国怎么办？_!_
+6552478165090959619_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552176401020092685_!_110_!_news_military_!_联合国安理会改革废除一票否决权，会导致什么后果？_!_
+6552436025153552899_!_110_!_news_military_!_控枪？拥枪？美国民众存在根本分歧_!_拥枪,达拉斯市,美国,美国步枪协会,得克萨斯州
+6552243771923235076_!_110_!_news_military_!_日本强硬向美说不，是真心还是作秀？_!_
+6552452522043245059_!_110_!_news_military_!_强硬报复！俄战机猛烈轰炸，支美武装据点全部摧毁，美军没有发声_!_俄罗斯,基地组织,叙利亚政府军,武装分子,哈马省
+6552475115022450952_!_110_!_news_military_!_美国近来不断扩张军力增加国防预算，恢复第二舰队，战略政策是扩张性的吗？_!_
+6552415051091804679_!_110_!_news_military_!_华州：军地联谊搭鹊桥，兵哥靓妹喜牵手_!_诗经,红一连,兵哥哥,大龄青年,军中绿花,缘来是你,军地,爱人同志,关雎
+6552453633097597454_!_110_!_news_military_!_叙利亚战争，美俄第一次军事冲突事件回顾！_!_叙利亚,俄罗斯雇佣兵,幼发拉底河,俄罗斯军队,美军
+6552450106510017037_!_112_!_news_travel_!_平邑：大麦滩的樱桃熟了_!_大麦滩,临涧镇,樱桃园,樱桃,平邑
+6552448065171620356_!_112_!_news_travel_!_越看越亲切的越南，不得不说的中西文化交融之地_!_越南,传统文化,北京紫禁城,胡志明市,文化
+6552459808945996295_!_112_!_news_travel_!_乌鲁木齐开展“民族团结一家亲”主题活动_!_沙汗,乌鲁木齐市,乌鲁木齐,艾尼瓦尔江,农家乐
+6552481541421793540_!_112_!_news_travel_!_焊不锈钢，铝，铜，用那种焊枪最好？_!_
+6552330349379060232_!_112_!_news_travel_!_浑江区有这样一个地方，每到六、七月份竟能美成这样……_!_青山湖,龙山湖,长白山,白山市,浑江区
+6552460893773365511_!_112_!_news_travel_!_世界第一伟人是谁？_!_
+6552445203490275597_!_112_!_news_travel_!_四川游客张某带着六名女子在三亚沙滩烧烤被撒沙，这事你怎么看？_!_
+6552350011236024580_!_112_!_news_travel_!_新买的iPhone7自带的地图里面为什么会有莫名其妙的导航记录？_!_
+6551675115115905544_!_112_!_news_travel_!_中国堪比比萨斜塔的塔，门票不到一百！周末就去这里了_!_虎丘,苏州,书法家,吕洞宾,虎丘塔
+6552387897545196040_!_112_!_news_travel_!_日本最环保的小镇：15年来零垃圾，一群老人靠捡树叶年入千万_!_德岛县,上胜町,株式会社,垃圾回收,环境保护
+6552395498110583310_!_112_!_news_travel_!_他单人单骑川藏80天：到达尼泊尔 回家后妈妈都认不出他了_!_人在野,背包客,日喀则,绒布冰川,尼泊尔
+6552467174777160196_!_112_!_news_travel_!_广东今年将打造11条南粤古驿道重点线路，初夏郊游走起~_!_海丰,平远,罗浮山,古道,南雄,连州,广东水利史,古驿道,乳源,广州府志,从化县志,寻乌调查,广东通志,明代驿站考
+6552298044140814861_!_112_!_news_travel_!_我和狗狗一起参加了登山比赛，我们完成了比赛，我们胜利了_!_狗狗
+6552490712225546766_!_112_!_news_travel_!_马略卡岛骑行的旅客指南—下_!_马略卡,马略卡岛,Major,自行车,骑行
+6552484625325752846_!_112_!_news_travel_!_坐着火车去拉萨，原来是这样的体验……_!_唐古拉山,格尔木,高原反应,西藏,拉萨
+6552482275647291662_!_112_!_news_travel_!_湘西那边到底有没有蛊？_!_
+6552476613282365965_!_112_!_news_travel_!_成都首座高空玻璃栈道五一开放，在哪？_!_玻璃,平乐镇,张家界,成都
+6552317903415280141_!_112_!_news_travel_!_洋眼看江苏活动走进大丰_!_江苏,洋眼看江苏,意大利,俄罗斯,大丰
+6552415206692094222_!_112_!_news_travel_!_在日本免税店被坑，买的东西要怎么退货？_!_
+6552347178877059342_!_112_!_news_travel_!_四川游客张某带着六名女子在三亚沙滩烧烤被撒沙，这事你怎么看？_!_
+6552313758851531271_!_113_!_news_world_!_最后一辆图坦卡蒙战车迁至大埃及博物馆_!_博物馆,图坦卡蒙,埃及博物馆,开罗,文物
+6552239578235798030_!_113_!_news_world_!_日本儿童人口连续37年减少_!_东京都,日本儿童人口,日本总务省,联合国,一级行政区
+6552419044442505732_!_113_!_news_world_!_特朗普为支持拥枪 拿伦敦当“反面教材”_!_恐怖分子,特朗普,得克萨斯州,达拉斯市,佛罗里达州
+6552305482256613645_!_113_!_news_world_!_特朗普是历任美国总统中最大的天才还是最大的蠢货？_!_
+6552285687591207428_!_113_!_news_world_!_谢丽尔·科尔：被认为是最美丽的英国女性，英国民众心目中的女神_!_
+6552300146858656269_!_113_!_news_world_!_俄罗斯如今左右为难，支援叙利亚受阻，只因影响到这个国家_!_防空系统,以色列,俄罗斯,S300,叙利亚
+6552346429963108615_!_113_!_news_world_!_退休后工人农民全部免费医疗，养老金统一，会怎么样？_!_
+6552269613172785668_!_113_!_news_world_!_战事升级！普京正式授权驻叙俄军“核武反击”!_!_苏联解体,叙利亚,以色列,普京,俄罗斯
+6552328646416138509_!_113_!_news_world_!_特朗普为什么要跟伊核协议较劲？_!_
+6552251982508196360_!_113_!_news_world_!_超越英国！全球第五大经济体不是国家而是这地区……_!_加州人口,GDP,经济体,国民生产总值,英国
+6552391967869764109_!_113_!_news_world_!_印度是真的落后，还是只是在中国游客的眼中，印度很落后？_!_孟买,印度,贫民窟
+6552345707112563203_!_113_!_news_world_!_俄将举行庆祝卫国战争胜利日阅兵活动，在彩排时它们也来凑热闹_!_莫斯科,普希金,纪念碑,普希金广场,东方IC
+6552444757530902788_!_113_!_news_world_!_如果现在是希拉里当总统，美国的外交政策会是什么样？_!_
+6552272967286391047_!_113_!_news_world_!_如果日本国土滑向海沟沉没，要求中国人民收留，作为中国人你会接受日本人吗？_!_
+6552455528855896324_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552347574567698696_!_113_!_news_world_!_巴基斯坦和伊朗会一起入侵以色列吗？_!_
+6552321762611167501_!_113_!_news_world_!_2018年美国中期选举，你认为特朗普会下台吗？_!_
+6552240226188656899_!_113_!_news_world_!_特朗普只是一个商人，更是一个政治素人，美国如此发达的国家为什么会选其当总统？_!_
+6552393914991510019_!_113_!_news_world_!_美国爸爸在网上分享了妻子的剖腹产账单，原来在美国生孩子这么贵_!_Reddit,账单,生孩子,剖腹产,妇女健康,约合
+6552378033334911235_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552376326253183492_!_115_!_news_agriculture_!_我国大量的进口的国外大豆，用到哪里去了呢？看完才明白！_!_美国,大豆,副产品
+6552130793114697988_!_115_!_news_agriculture_!_给植物喷施硫酸亚铁的浓度是多少才合适？_!_
+6552294626730967565_!_115_!_news_agriculture_!_稻种有缺陷水稻大减产_!_安徽省,合肥市,涉案稻种
+6552405592005149191_!_115_!_news_agriculture_!_小记——我记忆中最温暖的时光_!_农村城镇化
+6552383605727822343_!_115_!_news_agriculture_!_农村人常常吃的野菜，在专家眼里还有这样的功效！_!_农村,野菜,血皮菜
+6552371148917572104_!_115_!_news_agriculture_!_5月上中旬，小麦要高产还需怎么样管理？选择什么药剂？_!_病虫害,药剂,小麦,红蜘蛛,三唑酮
+6552427484879520264_!_115_!_news_agriculture_!_蟹塘水浑浊的原因及处理方案_!_水体,水草,黄颡鱼,混浊度,河蟹
+6552460614294307335_!_115_!_news_agriculture_!_4.5都有了！猪场下一步该怎么走？_!_新发地,养猪业,家庭猪场,猪价,专业育肥户
+6552363028841496835_!_115_!_news_agriculture_!_猪价持续低迷，影响猪价的因素有哪些，该如何扶持，才能减少养猪大户的亏损？_!_
+6552302857108848909_!_115_!_news_agriculture_!_东北经济的下一个增长点在哪里？_!_
+6552274062431748360_!_115_!_news_agriculture_!_母猪产仔咬仔猪怎么办？_!_
+6552273537967587854_!_115_!_news_agriculture_!_悄悄告诉你，如今不少果农选择这些“金牌”果树，果子清甜可口_!_土壤肥力,有机质,营养元素,链霉素,吲哚乙酸
+6552473582419575303_!_115_!_news_agriculture_!_养家糊口的农村妇女千千万，能把特产卖成王的只有麻城这一个_!_大别山,农产,中药材,父老乡亲,麻城特产
+6552483180442550787_!_115_!_news_agriculture_!_每袋135元的肥料最多值60多元！忽悠团联合村干部合伙坑害农民！_!_忽悠团,有机无机复混肥,甲壳素,化肥,村干部
+6552281735004946948_!_115_!_news_agriculture_!_观点：部分贫困群众各种“怕”是迟滞脱贫因素_!_羊肚菌,竹荪菌,贫困群众,贫困地区,顺势而为
+6552450087862141443_!_115_!_news_agriculture_!_江西“落单”白鹤恢复状态良好 明日乘飞机至吉林放飞_!_落单,爱爱,白鹤,江西省野生动植物救护繁育,鄱阳湖
+6552299652446683651_!_115_!_news_agriculture_!_茄果类高产第一步_!_茄果类,营养剂,营养
+6552308738613576195_!_115_!_news_agriculture_!_奇怪的现象，这个时代为什么只有粮食不值钱？_!_粮食价格,城里人
+6552374507544248840_!_115_!_news_agriculture_!_农村集市上的鱼，新鲜又便宜，白鲢是做麻辣鱼的最佳选择_!_麻辣鱼,农村,水煮鱼
+6552418785821721101_!_115_!_news_agriculture_!_「招人烦」的村第一书记_!_三变改革,五龙捧圣,第一书记,白河县,危房改造,双安村
+6552414017632076291_!_116_!_news_game_!_王者荣耀排位征召，非你们不ban_!_貂蝉
+6552313359310520580_!_116_!_news_game_!_《王者荣耀》的哪个英雄或者哪个技能最让你怀念呢？_!_
+6552109385831678472_!_116_!_news_game_!_《荒野行动》90%的人都不知道的钻石获取方式！不用太浪费了_!_赞赏值,钻石,荒野行动,游戏
+6552136830970495236_!_116_!_news_game_!_王者荣耀的哪个英雄或者哪个技能最让你怀念呢？_!_
+6552282323356746247_!_116_!_news_game_!_英雄联盟最秀的AD中单，男刀第三，第一秀到你怀疑人生_!_中单,W技能,对面,亚索,火药桶
+6552486585403703822_!_116_!_news_game_!_同等经济哪4个英雄最能抗伤害，白起难入榜，第一名没四个人打不死_!_夏侯惇,杀不死
+6552412702134764036_!_116_!_news_game_!_MSI入围赛Day4第一场：PGM 0-1 SUP，SUP再取一胜领跑B组_!_SUP,PGM,奥恩,卡尔玛,巨魔
+6552366221033996551_!_116_!_news_game_!_绝地求生5月份新版本都有哪些改动？_!_
+6552477550457651463_!_116_!_news_game_!_《王者荣耀》有哪些法师不依赖铭文？_!_
+6552488567233315075_!_116_!_news_game_!_《王者荣耀》有哪些法师不依赖铭文？_!_
+6552449708617368077_!_116_!_news_game_!_Uzi获得冠军后直播首秀730万人气，首局游戏老鼠拿下五杀_!_Uzi,MSI,艾欧尼亚,小狗,RNG签名队服
+6552339306956980743_!_116_!_news_game_!_DNF：玩家打团时睡着了，为了守护阿拉德大陆殚精竭虑_!_dnf,阿拉德,阿拉德大陆
+6552436963197059598_!_116_!_news_game_!_王者荣耀中这些情侣CP，你觉得哪对最虐汪？_!_虞姬,三姓家奴,小乔,凤凰于飞,吕布
+6552303102945395213_!_116_!_news_game_!_2018季中冠军赛第四日赛事前瞻：B组入围赛谁能觐见FW？_!_b组,GMB,PGM,大洋洲,土耳其
+6552354100946665735_!_116_!_news_game_!_如果迷你世界和Mc相互PK，你觉的谁会赢？_!_
+6552312485871878663_!_116_!_news_game_!_旅行青蛙中国之旅什么时候上线 中国之旅上线时间一览_!_青蛙中国,旅行青蛙,青蛙
+6552272739510518029_!_116_!_news_game_!_为什么《辐射4》被很多老玩家喷？_!_
+6552391553426391303_!_116_!_news_game_!_你玩的第一款网络游戏什么？_!_
+6552278657048510734_!_116_!_news_game_!_我的世界有多少个遗迹？种子是什么？_!_
+6552399691764465927_!_116_!_news_game_!_网易新游《第五人格》到底在玩什么？_!_
+6552127522933309703_!_101_!_news_culture_!_上联：一片彩霞迎旭日，怎么对下联？_!_
+6552455667020464654_!_101_!_news_culture_!_从《中庸》到《声律启蒙》，不一样的小朋友，一样的萌｜赵师汉学_!_小状元,中庸,赵师,宋就浇瓜,声律启蒙
+6552336898092695821_!_101_!_news_culture_!_鼻烟壶是什么？做什么用的？_!_
+6552444720486809870_!_101_!_news_culture_!_新时代武侠小说的开山鼻祖到底是谁？_!_
+6552360098616836365_!_101_!_news_culture_!_你心目中的伟人是谁？_!_
+6552318647426089219_!_101_!_news_culture_!_上联：院中红杏多浇水，下联怎么对？_!_
+6552337670561857799_!_101_!_news_culture_!_《西游记》最后为什么两个和尚要向唐僧他们收贿赂才肯把真经给他们？_!_
+6552288059243626760_!_101_!_news_culture_!_你见过最狂的一句诗词是什么？_!_
+6552413128812921095_!_101_!_news_culture_!_大三学生，最近书荒得厉害，有什么好看的书单推荐？_!_
+6552129543635730696_!_101_!_news_culture_!_古代的状元放到现在是不是大学生？_!_
+6552433617371398407_!_101_!_news_culture_!_什么是方法论？怎么掌握方法论？_!_
+6552447434755146243_!_101_!_news_culture_!_画家杨象乔以作品《蚩尤》讴歌民族精神_!_蚩尤,战神蚩尤,少数民族,中国神话,战神,中华民族
+6552414539290247437_!_101_!_news_culture_!_有哪些文学霸气句子？_!_
+6552403226354778381_!_101_!_news_culture_!_上联：清源山中鸟鸣喧，下联如何对？_!_
+6552342390286320135_!_101_!_news_culture_!_中国赠送马克思雕像在德国特里尔揭幕_!_特里尔,揭幕仪式,新华社记者,马克思雕像,单宇琦
+6552206417409868039_!_101_!_news_culture_!_诸葛亮能借东风火烧曹营，为什么火烧司马懿却没料到会下雨？_!_
+6552267795789250819_!_101_!_news_culture_!_上联：青山翠竹好时光，如何对下联？_!_
+6552182374606569741_!_101_!_news_culture_!_上联：连天风雨送四月，如何对下联？_!_
+6552128002312896772_!_101_!_news_culture_!_会龙二字作一对联，最好将二字分别放在上下联的头一字，如何作？_!_
+6552336664591597837_!_101_!_news_culture_!_你心目中的伟人是谁？_!_
+6552489357805093389_!_102_!_news_entertainment_!_女神来了！汤唯着浅V白裙秀美腿出席活动 大方挥手笑容温暖迷人_!_汤唯,白裙,女神
+6552311770151649796_!_102_!_news_entertainment_!_麦莉·赛勒斯去电视台录节目，她真是越来越美！_!_电视台,麦莉·赛勒斯,女歌手
+6552342421387084292_!_102_!_news_entertainment_!_瑞安雷诺兹分享死侍复仇者拒绝信_!_斯塔克,复仇者,死侍,无限战争,瑞安·雷诺兹,Maple
+6552377030304858631_!_102_!_news_entertainment_!_白敬亭奶凶登上热搜，可爱又帅气的白侦探，调皮起来也是又奶又凶_!_大侦探,我是大侦探,综艺,怀柔区,白敬亭
+6552350528137855502_!_102_!_news_entertainment_!_《后来的我们》都成长了 那么你懂得珍惜了么_!_后来,公务员,摸爬滚打,不名一文,那些年,北京人
+6552468411610300935_!_102_!_news_entertainment_!_出生到现在从来没有谈过恋爱？不可置信的母胎单身偶像TOP7_!_金所炫,Dawn,谈恋爱,母胎单身,吕珍九
+6552258492902670852_!_102_!_news_entertainment_!_《街舞》总决赛韩宇pk田一德夺冠，网友：可惜杨文昊battle没看够_!_田一德,这就是街舞,杨文昊,王子奇,舞者,总决赛
+6552168471738712584_!_102_!_news_entertainment_!_娱乐圈不吸烟的乖乖男，马天宇张艺兴上榜，他因不吸烟有不老容颜_!_黄金瞳,轩辕剑之汉之云,极限挑战第三季,奇幻冒险王,林志颖,剧中饰演,超级演说家第二季,霍去病,爸爸去哪儿,我对时间有耐心,三生三世十里桃花,马天宇,建军大业,张艺兴,迪丽热巴
+6552390995558793735_!_102_!_news_entertainment_!_刘德华的吃相、刘晓庆的吃相、徐峥的吃相，最后一个最丑让人恶心_!_赵薇,高晓松,徐峥,刘德华,刘晓庆
+6552452321333215752_!_102_!_news_entertainment_!_笑笑宣布离婚：人命关天，各位兄弟也别说啥了_!_毫无防备,抑郁症,重度抑郁症
+6552315418969637379_!_102_!_news_entertainment_!_史诗级奇幻大片《阿修罗》将于暑期上映，吴磊领衔剧照流出_!_琅琊榜,红海行动,捉妖记2,吴磊,封神榜之凤鸣岐山,王之一,电影制作人,阿修罗
+6552293881772245261_!_102_!_news_entertainment_!_怎么评价潘乱写的《腾讯没有梦想》？_!_
+6552455474866815491_!_102_!_news_entertainment_!_《创造101》黄子韬battle王一博，你更喜欢谁？_!_创造101,主题曲,黄子韬,王一博
+6552462629913231629_!_102_!_news_entertainment_!_到底娱乐圈有多难混？_!_
+6552391788840092173_!_102_!_news_entertainment_!_丁海寅C位被骂，对比金秀贤螃蟹步，网友群嘲：想红想疯了！_!_当你沉睡时,孙艺珍,丁海寅,金南珠,李相烨,金秀贤
+6552857959766950404_!_102_!_news_entertainment_!_林允骂哭同剧演员？金晨自爆恋情？大S吓跑投资人？_!_林允,演员,金晨,黄圣依,大s
+6552456830994350595_!_102_!_news_entertainment_!_谢楠带儿子现身，吴所谓一身潮装对镜卖萌，网友：一点都不像吴京_!_谢楠,战狼2,孤注一掷,镜卖萌,吴京
+6552445745092362765_!_102_!_news_entertainment_!_大侦探：导演为杨幂发声怒怼网友，但韩雪、鬼鬼、王嘉尔才更憋屈！_!_韩雪,明星大侦探,王嘉尔,诈捐门,杨幂,我是大侦探
+6552029422789067272_!_102_!_news_entertainment_!_对于生孩子这件事，黄磊的话让人尬，陈浩民渣，最暖心的是他！_!_黄磊,渣男,张晋,陈浩民,蔡少芬
+6552315678467031555_!_102_!_news_entertainment_!_李小璐商演捞金，依旧靠露肉博眼球全程面无表情心情不好_!_捞金,李小璐,贾乃亮
+6552386111505695246_!_103_!_news_sports_!_白送一个点球！恒大都不要？卡纳瓦罗笑容真有意思_!_卡纳瓦罗,朱晓刚,点球,恒大,大连一方
+6552314957298401539_!_103_!_news_sports_!_刘诗雯戴罪立功取胜平野美宇为中国女乒拿到世乒赛冠军，怎么看待这场比赛？_!_
+6552332692157891086_!_103_!_news_sports_!_天堂与地狱是有多近？看看比赛最后8秒的猛龙就知道了_!_猛龙,詹姆斯,步行者,凯尔特人,十之八九
+6552285072488137229_!_103_!_news_sports_!_鲁能队员的身高_!_足球,鲁能,王大雷,鲁能队,鲁能泰山队
+6552480288448971278_!_103_!_news_sports_!_WKG&M-1世界综合格斗赛明星见面会引爆鹏城_!_UFC,MMA,沃尔科夫,WKG,格斗迷
+6552421256900116743_!_103_!_news_sports_!_如何看待第一詹黑皮尔斯，公开赞美詹姆斯？_!_
+6552302468556915214_!_103_!_news_sports_!_粤媒：缺少外援的恒大还是不行_!_中超,里皮,国足,外交辞令,亚洲杯
+6552414720987496964_!_103_!_news_sports_!_热刺作死的节奏！连垫底队都输，欧冠席位悬了？_!_纽卡斯尔联,欧冠,热刺,西布朗,切尔西
+6552268881077993992_!_103_!_news_sports_!_德帅谈没球队面试布兹德里克：出于自私的原因，他在这让我开心_!_坐失良机,德里克,德安东尼
+6552349457025860110_!_103_!_news_sports_!_双刃剑！山东鲁能高中锋让人又爱又恨_!_李霄鹏,中超赛场,塔尔德利,佩莱,山东鲁能
+6552291533125583367_!_103_!_news_sports_!_克洛普：希望凯塔能迅速起到作用，可能会邀请他去基辅_!_欧冠决赛,莱比锡,基辅,凯塔,利物浦
+6552456441481921038_!_103_!_news_sports_!_在欧洲是夺欧冠难，还是夺联赛难？这篇文章将告诉您答案！_!_联赛冠军,夺得欧冠,欧冠冠军,欧冠夺冠,齐达内
+6552285814024306957_!_103_!_news_sports_!_斯诺克世锦赛半决赛，马克-威廉姆斯17-15霍金斯，希金斯和威廉姆斯谁能夺冠？_!_
+6552438970507067656_!_103_!_news_sports_!_脂肪肝怎样最快消失？_!_
+6552369842349933064_!_103_!_news_sports_!_韦德的一句bad boy！网友们炸了：个个都是段子手！_!_骑士队,韦德,猛龙队,猛龙,詹姆斯
+6552377102237171981_!_103_!_news_sports_!_中国队亚洲杯抽到上上签，为何里皮脸色仍然很凝重？_!_
+6552263064341709315_!_103_!_news_sports_!_今日直播龙星战及女子围甲 + 星阵让先9连胜 + 暗渡陈仓_!_龙星战,赵汉乘,古灵益,崔哲瀚,柁嘉熹
+6552281431614161165_!_103_!_news_sports_!_如果勇士今年夺得总冠军，是不是就算建立了王朝？_!_
+6552333218027143683_!_103_!_news_sports_!_横扫东部第一？这事骑士已经干过一次！直接把对手打到解体重建！_!_猛龙,季后赛,老鹰,东部决赛,科沃尔
+6552326316648038916_!_103_!_news_sports_!_2018陆家嘴金融城足球联赛昨天拉开帷幕_!_足球联赛,陆家嘴金融城足球联赛,金融城,循环制,海尔集团
+6552324349288775949_!_104_!_news_finance_!_车子安装了高速etc，并且绑定的是中行信用卡。现在车准备卖了，etc怎样处置？_!_
+6552469573004689934_!_104_!_news_finance_!_股东代表诉讼应定额收取诉讼费_!_三联商社,公司法,民事判决书,股东代表诉讼,联邦地区法院诉讼法,诉讼费,人民法院诉讼费用交纳办法,商法,共同诉讼,联邦民事诉讼法,实体法
+6552261707316592904_!_104_!_news_finance_!_余额宝有可能发行自己的银联卡，建自己的取款机吗？_!_
+6552414747973648909_!_104_!_news_finance_!_摆地摊小本生意卖什么最赚钱，那种销售模式适合你_!_盘发器,顺口溜,销售模式,小本生意,生意人
+6552399804914204935_!_104_!_news_finance_!_股票，期货配资有没有市场前景？_!_
+6552354112107708935_!_104_!_news_finance_!_重磅：涨停板复盘！_!_金刚玻璃,沙河股份,智光电气,华北制药,高送填权,星星科技,广联达,江龙,高斯贝尔,永泰能源,金运激光,云计算
+6552422565342609667_!_104_!_news_finance_!_目前有35万存款，怎样在短期之内翻倍？_!_
+6552353569004061197_!_104_!_news_finance_!_4月CPI、进出口等数据将公布，这些投资机会最靠谱_!_M2,皖维高新,云计算,江苏有线,乐普医疗,浪潮信息,逆回购,金龙汽车,东方雨虹,一带一路,奥瑞德,中信证券,Shibor,医疗旅游,CPI,净回笼,绿盟科技,Wind,新五丰
+6552259857167155460_!_104_!_news_finance_!_中国排在前600的姓都有哪些？_!_
+6552446825108865550_!_104_!_news_finance_!_大麻竟被做成了衣服_!_大麻,服装服饰,吉麻良絲,麻纺织,纺织
+6552436756438843908_!_104_!_news_finance_!_新政策新趋势，解读企业的上市背后的门道_!_内部控制,国金证券,上市,普华永道,IPO,生物医药,云启
+6552381165657915655_!_104_!_news_finance_!_两个人投资开店其中一个人不管，利润该怎么分？_!_
+6552404022110716420_!_104_!_news_finance_!_巴菲特，一生成功投资的最大秘密_!_现金流,巴菲特,虚拟货币,伯克希尔,股神
+6552463236896129543_!_104_!_news_finance_!_信用卡代还业兴起，众多平台被爆存违规套现、乱收费等问题_!_互联网金融,持卡人,信用卡代偿行业研究报告,信用卡,代偿,还款
+6552406300255322371_!_104_!_news_finance_!_比特币BTC和比特币现金BCH同时上涨，两者是否互相影响？_!_
+6552436333447479816_!_104_!_news_finance_!_军工行业11个世界500强大型企业集团资料整理_!_民用航空,歼击轰炸机,中国船舶,运-20,轰-6,航空工业,船舶工业,AC313,运载火箭,AC352,AG300,中船重工,AC311,中国兵器工业集团公司,歼10,中国航天,世界500强
+6552269056404095491_!_104_!_news_finance_!_如何判断热点题材板块的持续性？_!_涨停板,热点题材,持续性,指数走势,龙头股
+6552370609597186317_!_104_!_news_finance_!_很多散户资金量不多，却同时持有四五个，甚至十几个品种的股票，这样真的好么？_!_
+6552232276984332558_!_104_!_news_finance_!_现在手里面有两万块钱的闲钱，不是很多。这笔钱怎么在一年时间内赚最多的钱？_!_
+6552467065112887815_!_106_!_news_house_!_西安产融新时代 楼宇经济新方略_!_楼宇产业发展基地,陕西省,旭辉,西安市,研究会
+6552207942622380296_!_106_!_news_house_!_特朗普所做的一切会不会只是为了自己的商业帝国？_!_
+6552405108561281550_!_106_!_news_house_!_链家左晖：我们和同行早已不在一个时代_!_链家,供应链,经纪行业,经纪人合作,吃差价,格式化,不动产,经纪人
+6552272479094571277_!_106_!_news_house_!_燕郊房价还会涨到调控前的水平吗？_!_
+6552235099859976708_!_106_!_news_house_!_司法拍卖 东方太阳城 北京法拍房2万/平米在售 可贷款 可垫资_!_法拍房,东方太阳城,太阳城,桃花源,顺义区
+6552424582492455182_!_106_!_news_house_!_中国近20年的房地产高速发展，对中国科技的影响有多大？_!_
+6552319415294099971_!_106_!_news_house_!_房地产税改革渐行渐近_!_土地财政,瞭望,胡光辉,产权房,不动产,房地产税,税改革
+6552287375840510211_!_106_!_news_house_!_西安房价会涨过2万的均价么，如果跌的话还能跌回原来的6千么？_!_
+6552435689269494276_!_106_!_news_house_!_揭秘！为什么买新房找中介比去售楼处更便宜！_!_房价,找中介买新房,售楼处,更便宜,楼盘
+6552420718666055939_!_106_!_news_house_!_房价走势如何？_!_
+6552267646488805902_!_106_!_news_house_!_乌鲁木齐红山小区怎么样 好户型房价走势_!_楼盘,红山小区,房价,搜房网,红山
+6552343358537204231_!_106_!_news_house_!_买公寓，一看城市，二看回报率，除此之外都是没用的？_!_投资价值,写字楼,商住公寓,酒店式公寓,租金价值
+6552343170888237326_!_106_!_news_house_!_在西安买房哪个位置性价比比较高？_!_
+6552470047418221059_!_106_!_news_house_!_俸周耀：房地产品牌21条商规_!_物业,地产品牌,地产品牌建设,营销策划
+6552348127330828558_!_106_!_news_house_!_房价上涨买房想违约该怎么办？_!_
+6552491680002474254_!_106_!_news_house_!_关于要不要买房？_!_
+6552305949120397576_!_106_!_news_house_!_房屋被征收宅基有补偿吗？_!_
+6552290193473274126_!_106_!_news_house_!_家里有两个女儿，父亲不愿意买房子是什么心理？_!_
+6552265837334495757_!_106_!_news_house_!_市北区（青岛）80个小区最新二手房价，你知道的小区涨了？跌了？_!_小区市北区24,二手房价,小区市北区29,市北区,小区市北区25
+6552486788022141188_!_106_!_news_house_!_为什么说以后的城市空房会越来越多？_!_
+6552159026044469518_!_107_!_news_car_!_帝豪GL1.8自动精英裸车价86800，买贵了吗？_!_
+6552361234371445262_!_107_!_news_car_!_18款雷克萨斯LX570 凌志王者促销爆惠_!_SUV,语音识别,驾驶证,LX570,雷克萨斯LX570,雷克萨斯
+6552453721140232461_!_107_!_news_car_!_报警灯亮了刹车片就要更换吗？_!_
+6552130977726988552_!_107_!_news_car_!_2.4精英版欧蓝德换挡拨片怎么用？_!_
+6552443169743569412_!_107_!_news_car_!_中国车手何晓乐出征纽博格林24小时耐力赛，他喜欢什么样的耳机？_!_漫步者,Hybrid,赛车手,赛车,何晓乐,纽博格林
+6552369688507056388_!_107_!_news_car_!_2017新捷达1.5手动舒服型有哪些配置，买这车可以吗？_!_
+6552294558443504136_!_107_!_news_car_!_别还蒙在鼓里！这4款SUV就是换壳面包车，最后1辆月均2万人入坑！_!_CX70,宝骏,汉兰达,哈弗H6,长安CX70,五菱宏光,面包车,SUV,宝骏560,宝骏730,宝骏510
+6552365094741737987_!_107_!_news_car_!_大众集团提高电池交易量至480亿美元，或与特斯拉相同的市场容量_!_电动汽车,Mission E,保时捷,电动车,e-tron,大众集团,特斯拉
+6552248728978194952_!_107_!_news_car_!_跑完高速后，你的密集恐惧症犯了么？_!_清洁剂,黑胡子,洗车,挡风玻璃
+6552344274426396936_!_107_!_news_car_!_六万至八万元之间，买什么车比较好？_!_
+6552438515039207688_!_107_!_news_car_!_停车时先踩离合还是先踩刹车？_!_
+6552474103205331460_!_107_!_news_car_!_我一生的挚爱丰田皇冠汽车_!_操控性,发动机,雷克萨斯,丰田皇冠,宝马5系,皇冠,奥迪A6
+6552389610649944583_!_107_!_news_car_!_交警查车也有“潜规则”，他们就爱拦这些车，几乎一查一个准_!_电动车,半挂车,交通法,改装车
+6552287607005381123_!_107_!_news_car_!_最美中级轿车，配12.3英寸大屏，售16万，后悔买凯美瑞！_!_凯美瑞,豪车,标致,雅阁,标致508
+6552388591568290061_!_107_!_news_car_!_如何看待弯道超车的行为？_!_
+6552297756294119939_!_107_!_news_car_!_大众最得意B级车又降了！当年卖25万被疯抢，如今狂降9万没人买_!_迈腾,B级车,大众帕萨特,速双离合,凯美瑞,帕萨特,涡轮发动机
+6552371763823510020_!_107_!_news_car_!_大众得意B级车当年被疯抢，如今狂降9万没人买，家用还是很省心！_!_帕萨特,EA888,凯美瑞,额定功率,大众帕萨特
+6552373831862845965_!_107_!_news_car_!_来自未来的亮相！领先设计引领时尚新潮流_!_沃尔沃,领克01 PHEV,克02,能量晶体,免购置税新能源目录,新能源,PHEV
+6552271334422872583_!_107_!_news_car_!_女孩们 如何从“改车”看出另一半是暖男或渣男_!_行车记录器,改车,防倾杆,避震器,我家那口子
+6552311983197127176_!_108_!_news_edu_!_2018年高考时间确定！看2017年全国各省各批次录取分数线汇总！_!_高考分数线,民族班,理工类,普通班,文科,高考,招生考试,文史类
+6552491863360668163_!_108_!_news_edu_!_中考必考题之旋转（王老师每天一讲）_!_三角形,直角三角形,必考题
+6552320795308196109_!_108_!_news_edu_!_你怎么看待父亲精神分裂，母亲下肢残疾的他高考却被清华录取？_!_
+6552359134421844484_!_108_!_news_edu_!_侨报申请大学系列讲座 谷歌工程师林世彪主讲计算机专业与择校_!_谷歌,计算机专业,图形设计,林世彪,软件工程师
+6552448711329317128_!_108_!_news_edu_!_《瓦尔特保卫萨拉热窝》里的“劳费尔行动计划”是个什么计划？_!_
+6552032260818731533_!_108_!_news_edu_!_诸城市相州镇惠发小学强化学生感恩教育_!_诸城市,相州镇
+6552316460901859598_!_108_!_news_edu_!_苏州大学和东北财经大学哪个研究好？_!_
+6552297598735089927_!_108_!_news_edu_!_你见过最努力的人现在都变成什么样子了？_!_
+6552289670602949124_!_108_!_news_edu_!_即将中考的我_!_
+6552415459084337672_!_108_!_news_edu_!_教师招聘重要考点备考之中国古代教育专题_!_有教无类,孔子,九品中正制,科举制,察举制
+6552394201005293838_!_108_!_news_edu_!_在一个跟自己实力完全不符的学校学习是什么感受？_!_
+6552386523449262595_!_108_!_news_edu_!_「中考题型」物理想得满分？容易！只要你认真做这套题型！_!_题型,力学
+6552347851702141192_!_108_!_news_edu_!_你怎么看待父亲精神分裂，母亲下肢残疾的他高考却被清华录取？_!_
+6552178504329331208_!_108_!_news_edu_!_高校CP“秀恩爱”是怎样的体验，你想知道吗？_!_北大,清华学堂,清华,清华大学,工字厅
+6552336066601288200_!_108_!_news_edu_!_2天思维导图视频直播课+电子书+思维导图全套素材，全部免费送！_!_PMP,思维导图,墨尔本大学,鱼骨图,新加坡教育部
+6552326169117589773_!_108_!_news_edu_!_英语六级很难吗？_!_
+6552406113671708932_!_108_!_news_edu_!_你有哪些在外地求学/工作不会轻易向别人提起的心酸生活？_!_
+6552168301353500941_!_108_!_news_edu_!_微积分的本质是什么？_!_
+6552372630819701255_!_108_!_news_edu_!_北大校长致歉信近千字：只想让学生知道真实的自己_!_林建华,北大,鸿鹄,致歉信,未名BBS
+6552001414845956611_!_108_!_news_edu_!_2018中考英语最后冲刺书面表达之日记 记叙文or议论文_!_书面表达,English,adj.,中考英语,记叙文
+6552346329664717063_!_109_!_news_tech_!_在互联网创业需要什么？_!_
+6552421100343525895_!_109_!_news_tech_!_标题党真心严重，把腾讯坑坏了！_!_腾讯
+6552152270954299652_!_109_!_news_tech_!_高通骁龙845和835相比，性能有哪些提升？_!_
+6552407605774385416_!_109_!_news_tech_!_百度和谷歌的差距在哪里？_!_
+6552465915097645320_!_109_!_news_tech_!_三进制计算机可以实现吗？_!_
+6552297909570765315_!_109_!_news_tech_!_国内投资3000亿用来做芯片，够了么？能不能打败高通和英特尔？_!_英特尔,AMD,德州仪器,微处理器,高通
+6552458918214238477_!_109_!_news_tech_!_这个电脑配置值多少钱？_!_
+6552298922935582984_!_109_!_news_tech_!_美团众包以后会消失吗？_!_
+6552285961064022286_!_109_!_news_tech_!_新零售背景下，电商平台会发生什么变化？_!_
+6552281399263494404_!_109_!_news_tech_!_微信打不开了怎么办？_!_
+6552284076290605582_!_109_!_news_tech_!_腾讯发布“00后画像报告” 颠覆我们对这一代的认知_!_00后,00后画像报告,腾讯,大数据,腾讯QQ
+6551305201762959623_!_109_!_news_tech_!_持股比例仅次于雷军，如何评价小米二号人物林斌？_!_
+6552252696433262855_!_109_!_news_tech_!_玉米种子在播前，需要二次包衣吗？不需要或需要，又因为什么呢？_!_
+6552307609175589134_!_109_!_news_tech_!_淘宝购物大家能接受48小时后发货吗？_!_
+6552319719645380867_!_109_!_news_tech_!_电脑挂机软件是真的吗？_!_
+6552275262564729101_!_109_!_news_tech_!_索尼哪一款手机最好用？_!_
+6552415690236625421_!_109_!_news_tech_!_力哥爆料：如果有时光穿梭机，你最想见谁？马云想见的是他们_!_特拉维夫,以色列,阿里巴巴,马云,马克思
+6552479999771803912_!_109_!_news_tech_!_如果所有银行不让支付宝绑定银行卡了，支付宝会怎样？_!_
+6552320176207954190_!_110_!_news_military_!_军演时来敌人了怎么办？_!_
+6552210263775379720_!_110_!_news_military_!_女乒今天排兵布阵不合理，丁宁昨天刚打硬仗，今天应该打第一单打，你认为呢？_!_
+6552461852129886734_!_110_!_news_military_!_独步全球！沈飞又一款五代战机量产，性能让F35望尘莫及！_!_F35战机,歼31战机,沈飞,歼20,SR71黑鸟
+6552253268611825923_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552419215104540936_!_110_!_news_military_!_与歼20竞标的“雪鸮”为什么竞标失败？_!_
+6552386222843494925_!_110_!_news_military_!_观网小编进军营！全景展示74集团军_!_新西伯利亚,一支特别,网络名人进军营,硬骨头精神,74集团军
+6552282222043332878_!_110_!_news_military_!_一个月近900人死于枪击，拥枪对美国到底有什么意义？_!_
+6552444157137256708_!_110_!_news_military_!_美国声称“没有输过任何一场战役”，但是为何美国却输掉了越南战争？_!_
+6552454904122704391_!_110_!_news_military_!_五角大楼怒了！昨日两架美机硬闯我军基地，俩飞行员受伤紧急送医_!_21世纪海上丝绸之路,苏伊士运河,印度洋,非洲之角,五角大楼
+6552150320053485832_!_110_!_news_military_!_与恶狗对峙，该怎么办？_!_
+6552253467795128589_!_110_!_news_military_!_中国001航母和辽宁舰哪个强？_!_
+6552305665111490824_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552463827697402125_!_110_!_news_military_!_图-160战略轰炸机如今还是世界一流轰炸机吗？_!_
+6552443338820157700_!_110_!_news_military_!_美国航母即使被导弹打的千疮百孔也不会沉没，它采用的是什么钢铁结构船体？_!_
+6552489580996591880_!_110_!_news_military_!_叙利亚政府军缴获霍姆斯北的反对派军火库，发现大批军火是俄制，这是怎么回事？_!_
+6552410519033086215_!_110_!_news_military_!_叙利亚前线吃紧，北约舰队进入黑海，他们的目的是什么，你如何看待此时北约的军事行动？_!_
+6552236933139923463_!_110_!_news_military_!_中国此无人机直接对标美国“死神”，现在功能更强大使用也更经济_!_MQ-9,彩虹-4,有效载荷,彩虹-5,载弹量,导弹
+6552406297914901006_!_110_!_news_military_!_俄罗斯展示新型无人机，自称强于美军！不如买中国的吧_!_叙利亚,海盗,攻击机,俄罗斯,侦察机
+6552425203572408589_!_112_!_news_travel_!_咸阳和西安哪个城市的历史更悠久？_!_
+6552306441900786190_!_112_!_news_travel_!_大学毕业间隔年环游中国，做有意义的事情_!_间隔年,再见面,云南
+6552329949544448525_!_112_!_news_travel_!_看过了那些大山大水，初见阳朔竟是这般感受_!_喀斯特地貌,阳朔西街,啤酒鱼,阳朔,YSL
+6552416859419836932_!_112_!_news_travel_!_情报站｜在济南饮一碗“大碗茶”惬意舒适！_!_齐鲁壹点,李梅杰,趵突泉公园,趵突泉,大碗茶
+6552340425523331335_!_112_!_news_travel_!_洛阳和西安比，哪个城市历史更悠久？_!_
+6552332866305392900_!_112_!_news_travel_!_旅行时和异性拼房是一种怎样的体验？_!_
+6552480167128728067_!_112_!_news_travel_!_阆中嘉陵江两岸原生态自然风光，一个让人来了就不想离开的地方！_!_
+6552486879172755716_!_112_!_news_travel_!_重庆游玩有哪些旅游攻略？_!_
+6552421187232727300_!_112_!_news_travel_!_农村很多算卦的据说很准，你怎么看？_!_
+6552399922841256451_!_112_!_news_travel_!_海口塔昌村：打造原生态农宿 探索农业观光新业态_!_民宿,百香果,塔昌村,橡胶林,南海网
+6552454267284750851_!_112_!_news_travel_!_说桂林山水甲天下的人都是因为没有到过贵州，不信你看_!_毕节市,贵州省,游览,贵州,榕江县
+6552276552367735044_!_112_!_news_travel_!_旅行时和异性拼房是一种怎样的体验？_!_
+6552174212029612547_!_112_!_news_travel_!_骑车旅行到达大连晚上在海边别墅扎营_!_妈祖庙,大兴安岭,瓦房店,自说自话,海珠路
+6552245610378953220_!_112_!_news_travel_!_外媒称海南将成中国对外开放新窗口：繁荣或超香港_!_环岛高铁,海南省,自由贸易港,海南,拉美社
+6552301518295400974_!_112_!_news_travel_!_尼斯湖不仅有传说中的水怪，还有美不胜收的自然风景_!_尼斯河,厄克特城堡,水怪,尼斯湖,尼斯湖水怪
+6552309392237134350_!_112_!_news_travel_!_雷行记：来陕西旅游的朋友恭喜你，华山门票降价了！_!_华山,陕西,法门寺,门票,华清池
+6552358930171822340_!_112_!_news_travel_!_九月想自驾去西藏，如果我开车去，朋友坐我的车，费用怎么算才合理？_!_
+6552323082042737155_!_112_!_news_travel_!_广西再建城际，不走南宁，却连接了这两个穷地方，1300万人将脱贫了_!_梧州,梧州市,柳州,平南县,广西
+6552356661552480776_!_112_!_news_travel_!_中国最变幻莫测的公路：积雪比车高，一会下雨，一会下雪_!_丙察察线路,怒江,察隅县,怒江大峡谷,微言薄语
+6552468445223453187_!_112_!_news_travel_!_泰山酒店都不便宜，20年前都100多，现在的价格你肯定想不到_!_泰山,看日出
+6552353831185809668_!_113_!_news_world_!_作为俄罗斯总统，普京上台以来有哪些失误？_!_
+6552442168588370180_!_113_!_news_world_!_英国、俄罗斯、乌克兰、加拿大、澳大利亚、德国和西班牙，哪个国家的治安最好？_!_
+6552382245460508936_!_113_!_news_world_!_日本人如何夺走中国烹饪的世界地位？_!_
+6552420313383043336_!_113_!_news_world_!_美国封杀中兴通讯，不卖一颗芯片给中兴，究竟是为什么？_!_
+6552333666960277764_!_113_!_news_world_!_为何奥巴马刚上台就获诺贝尔和平奖？他给世界带来和平了吗？_!_
+6552470185742172424_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552335824208265476_!_113_!_news_world_!_为什么美国人一到18岁，马上会被父母赶出家门？_!_
+6552364621758464519_!_113_!_news_world_!_英国留下的种子发芽，东巴基斯坦独立的必然原因_!_东巴基斯坦,印巴分治,东巴,印度,西巴基斯坦
+6552368021996830989_!_113_!_news_world_!_哪个国家可能会成为下一个超级大国？_!_
+6552129532541796616_!_113_!_news_world_!_俄罗斯、土耳其、伊朗同盟已成，美国会对埃尔多安动手吗？_!_
+6552359592708276739_!_113_!_news_world_!_一篇很有意义的文章，希望您能耐心看下去_!_纪念馆,孔雀,鹅卵石,遇难者,孔雀园
+6552313247192580355_!_113_!_news_world_!_如果日本沉没，中国会接收日本难民吗？_!_
+6552142736877682952_!_113_!_news_world_!_有哪些科学解释不了的事实或现象？_!_
+6552363435361829389_!_113_!_news_world_!_印度和巴基斯坦为什么会仇视？难道仅仅是因为英国人的小手段吗？_!_印度教,印度,伊斯兰教,巴基斯坦,圣雄甘地
+6552227167177015559_!_113_!_news_world_!_如果美国打伊朗，最快可以多久解决战斗？_!_
+6552425808336519432_!_113_!_news_world_!_日本目前能修改和平宪法吗？_!_
+6552365932872728836_!_113_!_news_world_!_如果美国各个州独立了，世界格局会发生什么变化？_!_
+6552386615707173123_!_113_!_news_world_!_特朗普担任总统后，美国是更强还是更弱了？_!_
+6552452864088736014_!_113_!_news_world_!_非洲为什么没有熊？_!_
+6552260249565266445_!_113_!_news_world_!_俄罗斯90多个城市爆发反普京游行示威，警方逮捕上千人_!_纳瓦尔尼,普京,俄罗斯,反对派,欧洲人权法院
+6552288874138173960_!_115_!_news_agriculture_!_城里买房还不如回村建别墅，过幸福的田园生活_!_农村,别墅,浙江,田园
+6552266401208336904_!_115_!_news_agriculture_!_喜欢吃水果，可以考虑在家里自己种点，好吃还方便_!_山竹,山楂树,盆栽,嫁接苗,总状花序
+6552235829257830664_!_115_!_news_agriculture_!_什么品牌的矮壮素能防止玉米倒伏？使用防倒剂对玉米有何影响？_!_
+6552375956453982728_!_115_!_news_agriculture_!_「昌图牧歌」2018年5月6日猪价监测预警_!_牧歌,铁岭市,昌图县
+6552394108881601032_!_115_!_news_agriculture_!_“脱贫摘帽快 特色产业带”——祁东县精准脱贫记_!_中药材,农业,黄花菜,祁东县,五个十万
+6552336580483219972_!_115_!_news_agriculture_!_实在人做实在事_!_农村
+6552284559864496654_!_115_!_news_agriculture_!_天南地北怀化人│刘代全，精心打造文化“酉庄”_!_酉庄,文化,屈原,刘代全,九章·橘颂,尚书·皋陶谟,辰溪
+6552045667559145741_!_115_!_news_agriculture_!_如何高效防治葡萄的白粉病？_!_
+6552467307237474820_!_115_!_news_agriculture_!_猪场发生猪痢疾，给猪用这个方法治疗，1天控制死亡_!_猪拉血便,猪场,兽医,氟苯尼考,猪痢疾
+6552622022852084237_!_115_!_news_agriculture_!_玉米价格最好的上涨时机或许在6月下旬 秋季价格也不要太悲观_!_拍卖底价,玉米,中下旬,产玉米
+6552278972376285447_!_115_!_news_agriculture_!_稻苗长的慢可以用二胺吗？怎么使它快速生长？_!_
+6552437289421636110_!_115_!_news_agriculture_!_葡萄盆栽卖“景” 能吃能观赏收益超传统种植10倍_!_平度市,王桂欣,合作社,盆景葡萄,仁兆镇
+6552435081825223175_!_115_!_news_agriculture_!_蜂群失王后，工蜂还会采蜜吗？_!_蜂群,王台,蜂群失,幼虫,工蜂
+6552158165406843395_!_115_!_news_agriculture_!_9种好吃高营养的“特色”果树，趁早栽进地里，以后水果免费吃_!_山竹,树葡萄,果树苗,树苗,猕猴桃
+6552474741402239495_!_115_!_news_agriculture_!_2018年5月6日生猪价格行情 供应相对充足猪价逐渐趋于稳定！_!_外三元,猪价,浙江省,浙江,湖北
+6552321155754098951_!_115_!_news_agriculture_!_开始种植水稻了，秧田播种时，如何进行地膜保温育秧？_!_
+6552264023218323975_!_115_!_news_agriculture_!_立夏苗山农事忙_!_广西,广西柳州市融水苗族自治县,立夏,融水苗族自治县,杆洞乡
+6552338482365202701_!_115_!_news_agriculture_!_农村俗话“斗米养恩，担米养仇”应该怎样理解？_!_
+6552263170335965709_!_115_!_news_agriculture_!_止泻2小时见效就是这么简单，不信您试试！_!_腹泻,适应症,仔猪黄白痢
+6552378241703739651_!_116_!_news_game_!_2000左右买什么手机？_!_
+6552267362056274184_!_116_!_news_game_!_王者荣耀有哪些辅助可以出全输出装？_!_
+6552405780597833987_!_116_!_news_game_!_自从玩了《绝地求生》以后，你感觉到自己身体什么部位明显变差了？_!_
+6552029136498459149_!_116_!_news_game_!_《我的世界》攻略：手游地图生存指南，沙漠雨林海洋大盘点（上）_!_建议玩家,地图生存,可可豆,爬行者,指南针
+6552176715311874308_!_116_!_news_game_!_怎么看微博大V爆料王者掉帧卡顿等问题是因为手机品牌保护费交的不够？_!_
+6552462966879420936_!_116_!_news_game_!_LOL创造了吉尼斯纪录的英雄 第一是人面兽心 第三来自地狱_!_英雄联盟,千珏,吉尼斯纪录,游戏,LOL
+6552128061901373956_!_116_!_news_game_!_第五人格：厂长把傀儡扔上房顶后发现不对劲，游戏就这样结束了？_!_傀儡,求生者,监管者,厂长,傀儡扔
+6552437748089749774_!_116_!_news_game_!_你会为了《逆水寒》而放弃《剑网三》吗？_!_
+6552339621856936462_!_116_!_news_game_!_LOL：MSI投票成RNG战队秀？Uzi一个人把服务器给弄崩了_!_服务器,MSI,LPL,Uzi,LOL
+6552370661925323278_!_116_!_news_game_!_PS4独占大作《战神》结局深度解析_!_巴德尔,弗蕾雅,洛基,奎托斯,阿特柔斯
+6552130363123040771_!_116_!_news_game_!_DOTA2 英雄之路启程 MDL金鹰电竞站战队巡礼_!_战队,Major,Mineski,DOTA2,MDL
+6552402220178997512_!_116_!_news_game_!_有什么比较公平的网络游戏吗？_!_
+6552334261981020420_!_116_!_news_game_!_如果没有时间冷却，王者荣耀哪个英雄开大最厉害？_!_
+6552298932561510916_!_116_!_news_game_!_刺激战场：空投里不只是有三级头，还有AWM等各种狙击枪_!_狙击枪,绝地求生：刺激战场,M24,MK14,AWM,空投
+6515654933281243399_!_116_!_news_game_!_旅行青蛙能取代吃鸡吗？_!_团队游戏,游戏,竞技游戏
+6552297490048090631_!_116_!_news_game_!_王者荣耀：QG猫神：AG超玩会我们来陪你们了！_!_QG,职业选手,季后赛
+6552319000624235021_!_116_!_news_game_!_梦幻西游：这藏宝阁100万摆不开的召唤兽，10分钟之内必然被秒？_!_梦幻西游,藏宝阁,童子,召唤兽
+6552192763247460616_!_116_!_news_game_!_游戏开发公司有哪些岗位？_!_
+6531510403929735428_!_116_!_news_game_!_王者荣耀如果双方都不打，只让小兵上，谁会赢？_!_
+6551891535472886280_!_116_!_news_game_!_《DNF》玩家：玩了三个号实在有点玩不动了，该舍弃哪个号呢？_!_dnf
+6552256331951112462_!_101_!_news_culture_!_上联：对牛弹琴牛尾舞，如何对下联？_!_
+6552365635702096142_!_101_!_news_culture_!_上联：月下清贫花富贵，如何对下联？_!_
+6552441534225056013_!_101_!_news_culture_!_上联：雨润百草盛，如何对下联？_!_
+6552272366792081671_!_101_!_news_culture_!_今日立夏，有一上联，立夏万物并秀，下联怎么对？_!_
+6552241219840246280_!_101_!_news_culture_!_才子的行书这样美，没有不学的理由_!_文衡山,文征明,甫田集,岁贡生,朱宸濠,以字行
+6552393516239028488_!_101_!_news_culture_!_在晴雯不改变自己性格的情况下，如何改才能变自己的命运？_!_
+6552412813887799821_!_101_!_news_culture_!_现藏于台北故宫博物院的褚遂良《倪宽赞》欣赏_!_台北故宫博物院,倪宽赞,杨士奇,书法家,褚遂良,詹景凤
+6552344411164901639_!_101_!_news_culture_!_赵姨娘作为良妾，为什么在贾府的地位还没袭人高？_!_
+6552130055651197447_!_101_!_news_culture_!_立夏：春红落尽，夏木成荫_!_安家乡,黄莺儿,石榴花,夏日长
+6552437851689058573_!_101_!_news_culture_!_如何看待曹德旺说的，“一百对夫妻中，没有一对是幸福的”这句话？_!_
+6552353398312665352_!_101_!_news_culture_!_上联：桃花伴柳江湖远，如何对下联？_!_
+6552285150137286916_!_101_!_news_culture_!_上联：风吹干了相思忆，如何对下联？_!_
+6552180773846253828_!_101_!_news_culture_!_上联：雨淋荷花荷更洁，怎么对下联？_!_
+6552237111347511566_!_101_!_news_culture_!_上联：江中鱼东游西藏，下联怎么对？_!_
+6552341422786216196_!_101_!_news_culture_!_上联：月印荷塘荷印月，如何对下联？_!_
+6552187378222498307_!_101_!_news_culture_!_中国赠送马克思雕像在德国特里尔揭幕_!_马克思主义,莱法州,雕像,马克思,特里尔
+6552242505021128967_!_101_!_news_culture_!_上联：风伴红叶戏秋水，下联如何对？_!_
+6552421550627225863_!_101_!_news_culture_!_为什么电影《倩女幽魂》中燕赤霞一个道士却拿着金刚经？_!_
+6552400081205592583_!_101_!_news_culture_!_世界上最早的一副水彩画：丢勒《一大块草皮》_!_男人浴室,木刻版画,木版画,浪荡子,小受难,启示录,丢勒,版画家,亚当与夏娃,一大块草皮,水彩画,海怪,骑士、死亡与恶魔,基督大难
+6552486612830257416_!_101_!_news_culture_!_上联:今日头条，头条今日，说今日头头是道，论头条条条精华；如何对下联？_!_
+6552372964371726855_!_102_!_news_entertainment_!_最强组合！华晨宇合作莫文蔚，全是《天籁之战》牵线_!_半生缘,Karen,天籁之战,慢慢喜欢你,新声代,华晨宇,莫文蔚
+6552458141718544904_!_102_!_news_entertainment_!_作为湖北的家人，你知道从湖北走出的明星有哪些吗？_!_琅琊榜,丑女无敌,北京电影学院,中央戏剧学院,袁姗姗,北平无战事,宫锁心玉,少年杨家将,欢喜婆婆俏媳妇,美人心计,唐宫美人天下,李宗翰,金粉世家,宫锁珠帘,神雕侠侣
+6552332609655931400_!_102_!_news_entertainment_!_吴宣仪是孟美岐的小公主 李子璇受《创造101》剪辑偏爱_!_偶像练习生,撑腰,创造101,李子璇,吴宣仪,加油！美少女,Sunnee
+6552050834492883463_!_102_!_news_entertainment_!_盘点明星大佬们身旁都有些什么样的保镖_!_刘诗诗,吴奇隆,护花使者,刘亦菲,电影
+6552401595005403662_!_102_!_news_entertainment_!_后来的我们，不好不坏，你可以从任何方面批评，但它三观没有不正_!_拍电影,刘若英,张一白,三观不正,张嘉佳
+6552141844317209102_!_102_!_news_entertainment_!_《这就是街舞》真的有黑幕！从哪些地方可以看出？_!_这就是街舞,易烊千玺,王子奇,李琰,韩庚,黄景行
+6552387240520057351_!_102_!_news_entertainment_!_张靓颖自曝右耳间歇性失聪，演唱会上五字讲述未来让无数粉丝落泪_!_华语乐坛,粉丝,张靓颖,歌手,演唱会
+6552142812417098247_!_102_!_news_entertainment_!_奚梦瑶的腿怎么上热搜了？_!_维密,奚梦瑶维密超模奚梦瑶维密,超模奚梦瑶维密超模奚梦瑶,奚梦瑶
+6552377743093268995_!_102_!_news_entertainment_!_71岁施瓦辛格心脏手术后归来，与44岁女友双双现身_!_好莱坞,施瓦辛格,终结者
+6552460922571457027_!_102_!_news_entertainment_!_同父同母不同命，哥哥退出演艺圈生活窘迫弟弟曾红的发紫_!_吴启华,妙手仁心,倚天屠龙记,镭射青春,程至美,流氓大亨,慕容白,九品芝麻官,爱情全盒,魔域桃源,方唐镜,美味情缘,大运河,边城浪子,生死讼
+6552420340448887300_!_102_!_news_entertainment_!_EXID率智眼球开刀首露面！抗病1年近况曝光⋯粉丝哭了_!_Instagram,演唱会,instagram,EXID,粉丝
+6552408328486519309_!_102_!_news_entertainment_!_雷佳音刘昊然这么多相似地方？刘昊然模仿雷佳音，昊然弟弟你变了_!_郭京飞,佟丽娅,刘昊然,雷佳音,李光洁
+6552449197319127560_!_102_!_news_entertainment_!_跨界歌王薛之谦穿着时尚亮相，唱了一首绅士，开口全场都沸沸扬扬_!_薛之谦,综艺,歌王
+6552446812953772551_!_102_!_news_entertainment_!_谢娜创两项吉尼斯记录，中国第一人，网友回复最扎心_!_快乐家族,粉丝,快本,快乐大本营,谢娜,何炅,吉尼斯
+6552488595511312899_!_102_!_news_entertainment_!_本年度终身成就奖居然是他，和蔼可亲貌不惊人的老头_!_七十二家房客,唐山大兄,香港电影金像奖颁奖典礼,楚原,武侠电影,最佳男配角,孔雀王朝
+6552438994104222215_!_102_!_news_entertainment_!_angelababy 一起去海边吹吹风「污」男神女神一起驾到_!_驾到angelababy,男神女神,女神,angelababy,海边吹吹风
+6552343826801885710_!_102_!_news_entertainment_!_今31岁身家过亿，是赵丽颖的绯闻男友_!_李易峰,关晓彤,王一博,天天向上,赵丽颖
+6552423917078708743_!_102_!_news_entertainment_!_小奶狗撩妹套路深，《北京女子图鉴》陈可“姐弟恋”再陷感情危机_!_北京女子图鉴,小奶狗,陈可,奶狗,朱朝阳,杨大赫
+6552444286133076487_!_102_!_news_entertainment_!_日本选出全球最标致的五大面孔，中国上榜一人，不是范冰冰而是她_!_中国上榜一人,日本选出全球,范冰冰,五大面孔,标致
+6552314050561180174_!_103_!_news_sports_!_绝杀猛龙之夜！詹姆斯5次重摔在地，再心酸也值得了！_!_猛龙,詹姆斯
+6552484745748414979_!_103_!_news_sports_!_哦！中国第114个世界乒乓球冠军——王楚钦！为中国对加油！_!_王楚钦,乒乓球
+6552347256006115587_!_103_!_news_sports_!_朱婷在女排欧冠半决赛中受伤，这会对瓦基弗产生怎样的影响，还能够夺冠吗？_!_
+6552276948142260487_!_103_!_news_sports_!_德国男乒3:2艰难战胜韩国，挺进决赛会师国乒，中德乒乓大战胜算大吗？_!_
+6552385003618042115_!_103_!_news_sports_!_詹姆斯现在的绝杀球能力在联盟中是什么水准？_!_
+6552401077428290056_!_103_!_news_sports_!_淘汰恒大也保不住曼萨诺，保级战酿1-5惨案，球迷怒吼下课！_!_惠家康,足协杯,恒丰,泰达,曼萨诺
+6552373693555671566_!_103_!_news_sports_!_曝限薪令最快在夏窗出台 恒大重庆或成最大受害者_!_中超,限薪令,广州恒大,伊涅斯塔,袁野
+6552409524739768840_!_103_!_news_sports_!_58同城中甲：下半场的卓尔惹不起 他们这一数据冠绝联赛_!_卓尔,梅州客家,浙江绿城,佳兆业,武汉卓尔
+6552199075507208456_!_103_!_news_sports_!_瑞典世乒赛中国女队夺冠，如何看待新闻发布会上石川佳纯充当中国翻译？_!_
+6552489660398960910_!_103_!_news_sports_!_世乒赛男团决赛，中国队3-0完胜德国，两队的排兵布阵有哪些亮点和失误？_!_
+6552420043697684999_!_103_!_news_sports_!_道义放两旁利字摆中间，屡战屡败的曼萨诺还能走多远？_!_天津泰达,曼萨诺,贵州恒丰,文筱婷,北京国安
+6552318519168467213_!_103_!_news_sports_!_如果骑士今年一轮游，那么詹姆斯的历史地位会受影响吗？他还有机会再进总决赛吗？_!_
+6552447825676861699_!_103_!_news_sports_!_通过运动健身，你觉得对你最大的改变是什么？_!_
+6552220311318692100_!_103_!_news_sports_!_上港主场1-2负于国安，遭遇联赛两连败，上港到底输在哪？_!_
+6552473566724489732_!_103_!_news_sports_!_法网卫冕冠军在马德里首轮出局，法网卫冕前景蒙上阴影_!_罗马尼亚,法网,马德里,奥斯塔彭科,WTA
+6552362236168372494_!_103_!_news_sports_!_骑士和步行者战成抢七险胜，而猛龙即将被横扫，步行者跟猛龙的区别在哪里？_!_
+6552276966609781252_!_103_!_news_sports_!_中超单场消息：埃尔克森建功难救主 上港主场遭遇国安逆转_!_埃尔克森,国安,国安队,颜骏凌,上港队
+6552394344232386824_!_103_!_news_sports_!_惠若琪被问女排中素颜谁最美，她机智地回答郎平，你怎么看？_!_
+6552779221641462280_!_103_!_news_sports_!_维亚利：尤文不应该签埃姆雷-詹，而该选拉基蒂奇_!_维亚利,尤文,埃姆雷-詹,尤文图斯,拉基蒂奇
+6552137560229937422_!_103_!_news_sports_!_大连一方3:0完胜恒大，足协杯出局、又被一方大败，恒大怎么了？_!_
+6552357020526182919_!_104_!_news_finance_!_用超过1亿元的资金回购公司股份，这3家公司是怎么了？_!_威腾,超过,回购公司,回购股份,人民币
+6552369051719434766_!_104_!_news_finance_!_联想五大优势构建直面万亿市场的新业务模式SIoT_!_联想SIoT合作社,智慧联想,合作伙伴,云计算,物联网
+6551952253933060612_!_104_!_news_finance_!_宋清辉：资管新规出炉意味着打破刚性兑付_!_债券,中金公司,WIND,神雾环保,公司债,富贵鸟,洲际油气,违约
+6552283905750204686_!_104_!_news_finance_!_上证指数持续围绕3100点震荡，是筑底还是持续下跌？_!_
+6552450876856861192_!_104_!_news_finance_!_股票书籍分享——技术分析之五档买卖盘，还要练习盘感_!_股价,主力,卖盘,操纵股价,回调,西部矿业,大单,买盘,罗牛山,巨量,郑州煤电,量峰
+6552353187574055181_!_104_!_news_finance_!_以茅台酒现在的价格，你认为它还是用来喝的吗？_!_
+6552336506638303752_!_104_!_news_finance_!_币圈事件：密货币是否成为证券，5月7号见分晓，后市操作见下文_!_美国证券交易委员会,区块链,CFTC,SEC,梭哈
+6552428773889802503_!_104_!_news_finance_!_如何才能享受8%的定存利率？_!_
+6552400151267246606_!_104_!_news_finance_!_万孚生物：科金控股计划3个月内减持1%股份_!_万孚生物
+6552435222346990094_!_104_!_news_finance_!_蓝星生财之道公众号增长计划0.2版本_!_蓝星生财之道,公众号,区块链,北极星指标,核心价值,增长黑客
+6552277110025617667_!_104_!_news_finance_!_连续29个跌停刷新A股记录的ST保千里打开跌停，可否抄底？_!_
+6552357387150295566_!_104_!_news_finance_!_投资期限对资产配置的影响_!_股票投资,资产配置,压力测试,股票,投资期限
+6552315007487443469_!_104_!_news_finance_!_向初梵；5.6周一黄金操作建议，下周黄金走势分析_!_里奇蒙德,FOMC,发表讲话,亚特兰大联储,黄金,FOMC票委
+6552276057708298766_!_104_!_news_finance_!_网贷逾期后，既不催收也联系不到客服，却遭遇一天200元催收费_!_贷款人,网贷平台,催收费,网贷,逾期
+6549399653836653060_!_104_!_news_finance_!_三无人员申卡难，掌握申卡技巧快速下卡！_!_信用卡,申卡,持卡人,三无人员,民生银行,银行主推
+6552394200065769988_!_104_!_news_finance_!_有点闲钱了 放在哪里收益最划算_!_理财产品,银行理财,余额宝,流动性,货币基金
+6552392547614851335_!_104_!_news_finance_!_新手应该如何系统学习投资？_!_
+6552137748512244232_!_104_!_news_finance_!_巴菲特：不太可能向其股东发放一次性特别股息_!_巴菲特,股东
+6552379181982810638_!_106_!_news_house_!_小产权？公寓？没名额该如何买房？_!_律师见证书,合作建房转让协议,土地增值税,土地使用证,小产权,产权房,建筑许可证,LOFT,商务公寓,集资建房协议
+6552417626402849038_!_106_!_news_house_!_在深圳，作为有钱人是一种什么样的体验？_!_
+6552410089330835725_!_106_!_news_house_!_想去三四线城市定居，首付有限，是先买个偏点儿的房子，还是再攒几年钱后买好地段的房子？_!_
+6552381772514984461_!_106_!_news_house_!_美国富人的那些豪宅_!_豪宅居住面积,豪宅,内华达州,美国
+6552337461551300871_!_106_!_news_house_!_房子十年后会涨十倍吗？_!_
+6552293303411278083_!_106_!_news_house_!_大连东港看起来那么现代，为何房子卖不动？_!_
+6552297219511288068_!_106_!_news_house_!_新手买房，是找中介好还是自己去售楼部买更好？有什么需要注意的？_!_
+6552288840403386631_!_106_!_news_house_!_房价会不会再次上涨？_!_
+6552210325582643470_!_106_!_news_house_!_任志强发声：3月以后楼市肯定发生变化，你怎么看？_!_
+6552456242298618115_!_106_!_news_house_!_郑州哪家卤肉最好吃？_!_
+6552441742690353411_!_106_!_news_house_!_买房的时候五证不全，该怎么办？_!_
+6552364624103080200_!_106_!_news_house_!_现在在重庆投资房产还有升值空间吗？_!_
+6552360099904504333_!_106_!_news_house_!_3月房价下跌6元后 深圳4月跌16元又把大家逗笑了_!_深圳,网签备案合同价,资源委员会,评估价,二手房
+6552429611047715335_!_106_!_news_house_!_武汉：开发商不得将“房查”作为购房的前置条件_!_房产信息,开发企业,房管部门,房查,购房资格
+6552406240339689997_!_106_!_news_house_!_重庆刚需5月买房，能买洋房何必高层？_!_重庆楼市,高层,刚需,重庆,楼市
+6552461845842624772_!_106_!_news_house_!_山东日照2018年买了房子之后，房价跌了怎么办？_!_
+6552487566308802819_!_106_!_news_house_!_房价涨那么高，为什么还有人抢着买？_!_
+6552342374251495693_!_106_!_news_house_!_西安房子2018年能入手吗，会不会成接盘侠？_!_
+6552254215127826701_!_106_!_news_house_!_济宁市太白湖区碧桂园中央公园可以买吗？_!_
+6552400153234375176_!_106_!_news_house_!_山西：一季度商品房待售面积比上年末减少148万平方米_!_山西省,商品房销售面积,房地产市场,房地产,季度山西省商品房
+6552284465035477507_!_107_!_news_car_!_别说你很懂！踩刹车也是有技巧的_!_司机,踩刹车
+6552410665888252424_!_107_!_news_car_!_车主必须要了解的汽车玻璃知识_!_汽车玻璃,挡风玻璃,钢化玻璃,夹层玻璃,玻璃
+6552387385185796611_!_107_!_news_car_!_你的爱车装上这些小用品，既精美又显档次_!_
+6552306140342911239_!_107_!_news_car_!_抵押车要怎样年审和购买保险？_!_
+6552336549084660231_!_107_!_news_car_!_18款福特猛禽F150 3.5T白金版墨版皮卡大咖_!_福特F150墨版,越野车,福特,福特F150白金版,F150,F150墨版,福特F150
+6552350662745653763_!_107_!_news_car_!_自驾游不用住旅馆，新式“房车神器”横空出世！长这大还真头次见_!_拐角单筒望远镜,望远镜,便携式,扫地机,自驾游
+6552348739242033667_!_107_!_news_car_!_刚买H6就后悔了，猎豹新车比奥迪漂亮，搭宝马1.6T动力，仅11万起_!_玛莎拉蒂,奔驰S级,CS10,迈途,猎豹,CS9
+6552310231169237507_!_107_!_news_car_!_月工资多少才可以养得起10万左右的车子？看完后涨知识了_!_车子,月工资,4s店,司机
+6552410154996859405_!_107_!_news_car_!_10万元体验驾驶乐趣？看2016款新桑塔纳1.6L手动舒适版_!_手动挡,EA211,卤素大灯,桑塔纳,1.6L,新桑塔纳
+6552125903546106372_!_107_!_news_car_!_不管是奔驰，宝马，还是国产汽车，装上这些黑科技神器，高级有面_!_菲林片,音频线,车载空气净化器,车载充气泵,黑科技
+6552402268069560580_!_107_!_news_car_!_雅阁，凯美瑞，天籁后期费用，大约都多少钱？_!_
+6552413624088920589_!_107_!_news_car_!_MPV也加价？这款7座MPV，3米2 轴距2.0T比GL8还帅，还要GM8干啥_!_东风风行CM7,CM7,别克GL8,风行CM7,GL8,东风风行,GM8
+6552484499102368259_!_107_!_news_car_!_高尔夫GTI TCR公路版预告 5月底发布_!_高尔夫GTI,高尔夫gti,tcr,TCR,新车,高尔夫
+6552285933536805379_!_107_!_news_car_!_2年0息丨“万元”五重礼WEY你而来_!_SUV
+6552317337951797773_!_107_!_news_car_!_机械式销售——内饰清洗这样讲能多卖_!_施工单,内饰清洗,内饰
+6552349641625567757_!_107_!_news_car_!_长沙张公岭考场「科目三」直线行驶6大秘籍—恒哥话驾陪_!_参照物,张公岭,直线行驶,调整方向
+6552446289915675149_!_107_!_news_car_!_完美7座车和垃圾7座车，只差这么几样小东西_!_丰田,奥德赛,林肯领航员,福特,沃尔沃,比亚迪,锐界,儿童座椅,指挥官,汉兰达,Jeep,沃尔沃XC90,夏朗,撼路者,XC90,普瑞维亚,大众夏朗,D90,CS95,脚踏板
+6552304072588788227_!_107_!_news_car_!_3款被公认贬值最快的合资车，二手车贩都不愿意收_!_马自达6,福特,标致308,蒙迪欧,二手车,福特蒙迪欧
+6552478216219525645_!_107_!_news_car_!_全新外观/动力提升！新款Macan谍照曝光_!_Panamera,新款Macan,双涡轮增压,Macan,Turbo
+6552368521756541187_!_107_!_news_car_!_电动汽车是否真的能成为下一个汽车研究热点？_!_
+6552420505788350983_!_108_!_news_edu_!_全国首个人工智能课程走进中小学课堂_!_人工智能实践教程,教育信息化2.0行动计划,创造栗,人工智能,编程,新一代人工智能发展规划,朝阳区,初识语音识别,人工智能科普教育
+6552435876419338509_!_108_!_news_edu_!_高考有多累？_!_
+6552416572252619271_!_108_!_news_edu_!_我省2018年大学生志愿服务西部计划启动_!_公务员,陕西省,西部计划,大学生志愿服务西部计划,国家公务员
+6552461219817587203_!_108_!_news_edu_!_娄底城区2018年中职招生工作协调会在市第一职业中学召开_!_教育厅,招生,中职招生,职业中学,市教育局
+6552313284630937860_!_108_!_news_edu_!_读书是如何改变一个人的命运的？_!_
+6552438040134942989_!_108_!_news_edu_!_马上21岁了，一事无成，想学个技术。学什么好？_!_
+6552440328412987911_!_108_!_news_edu_!_冶师附小召开“家校携手育英才”家庭教育专家报告会_!_家庭教育,家长会
+6552276958112121348_!_108_!_news_edu_!_世界上最“神”速算法，秒杀计算器！只发一次，替孩子收藏！_!_计算能力,速算法,数学
+6552387448402346247_!_108_!_news_edu_!_女孩子去技校学什么专业比较好呢？_!_
+6552391519968428558_!_108_!_news_edu_!_零基础学摄影，轻松考大学，聊城高二学生的机会来啦_!_因材施教,摄影专业,影视摄影,聊城,南京艺术学院
+6552435721532080387_!_108_!_news_edu_!_高考要来了，高考回忆有哪些？_!_
+6552216062660706564_!_108_!_news_edu_!_你对哪些世界未解之谜感兴趣？_!_
+6552139666450022663_!_108_!_news_edu_!_从东京女留学生遇害案，说说清末为何喜欢扎堆留学日本？_!_
+6552373349480137230_!_108_!_news_edu_!_小学生策划消防教育活动 提升对消防知识的了解_!_消防,幼儿园,消防知识,小小消防员,史家小学
+6552342375228768520_!_108_!_news_edu_!_马上到美国留学的大学男生，对象找哪的合适？_!_
+6552360143063876103_!_108_!_news_edu_!_永定区：着眼教育管理理念 提升教育创新能力_!_业务校长,中小学校长培训规定,张家界市,永定区,教师
+6552132339051266563_!_108_!_news_edu_!_英语一天一句跟我读（第36天）_!_卷舌音,国家旅游局,Would,疑问句
+6552313849574326542_!_108_!_news_edu_!_流浪狗扰乱正常教学秩序，破坏学生们生活，该怎么办？_!_
+6552364182048604680_!_109_!_news_tech_!_2018-05-06 区块链中，代码即是法律？_!_EIP,内生规则,去中心化,互联网协议,比特币
+6552282943765610755_!_109_!_news_tech_!_邻居上门问wifi密码，怎么有技术又不伤感情地拒绝？_!_
+6552357712796057863_!_109_!_news_tech_!_现在最便宜的64G内存手机是1299元，未来10年，64G内存的手机能便宜到多少钱？_!_
+6552281225858384388_!_109_!_news_tech_!_未来有一半工作被取代，文科生最吃香！李开复给父母们的3个建议_!_异类,数位化,未来Family,AI,机器人,李开复,人工智能
+6552452862058692872_!_109_!_news_tech_!_小米所有的产品，有哪一种的核心技术是小米自己研发的？_!_
+6552349950380868110_!_109_!_news_tech_!_为了微信支付安全，请打开微信钱包这个功能吧_!_钱包锁,微信,微信钱包时,安全性,微信钱包
+6551669263533342989_!_109_!_news_tech_!_小米上市后市值若超过百度，会带来哪些影响？_!_
+6552460753587142920_!_109_!_news_tech_!_我想买探界者，探界者和昂科威一样的轴吗？_!_
+6552390444066537987_!_109_!_news_tech_!_马云放大招，网商银行转账全免费，马化腾腾讯微众银行如何应对_!_金融服务,支付宝,马化腾,微粒贷,微众银行
+6552466016436224526_!_109_!_news_tech_!_支付宝又有重榜好消息，4.5亿用户终于迎来了春天_!_余额宝,马云,支付宝,天弘基金
+6552362958964392200_!_109_!_news_tech_!_假如有个好产品，如何找一个好的营销团队？_!_
+6552257236930920718_!_109_!_news_tech_!_为什么台积电一个代工企业利润率在中国只输工行，还比苹果高呢？_!_
+6552477294743519491_!_109_!_news_tech_!_有哪些好用的护发产品？_!_
+6552458063855485447_!_109_!_news_tech_!_中科院发布一好消息，首款云端人工智能芯片问世，打破世界纪录_!_寒武纪,科大讯飞,机器学习,寒武纪1A,人工智能芯片
+6552437997374013709_!_109_!_news_tech_!_小米在港上市之后，雷军身价有多高？_!_
+6552326032085484039_!_109_!_news_tech_!_海尔总裁周云杰：中国智慧物流品牌日为世界贡献中国智慧_!_拥抱物联网时代,物联网,物联网时代,云计算,周云杰,生态系统
+6552495157961294083_!_109_!_news_tech_!_企业管理软件系统哪个好？_!_
+6552301085585834248_!_109_!_news_tech_!_手机电量显示100%，突然自动关机显示没电了，是什么原因？_!_
+6552390986826252803_!_109_!_news_tech_!_朱啸虎姚劲波……互联网大佬纷纷表态《腾讯没有梦想》_!_腾讯没有梦想,马化腾,朱啸虎,刘炽平,姚劲波,腾讯
+6552290024585429518_!_110_!_news_military_!_一个细节证明，B21轰炸机要服役了，全球没一个系统可以拦截_!_战略轰炸机,轰炸机,B-21,爱德华兹空军基地,B21
+6552259671955079694_!_110_!_news_military_!_沈飞歼31正密集试飞，复合材料比歼20还高5%，隐身性更强_!_F-22,歼-31,歼-20,F-35,战斗机,苏-27,SR-71,五代机
+6552271155430949127_!_110_!_news_military_!_美军控制区内出现新的化武袭击，俄罗斯为何保持克制不去揭穿？_!_
+6552347694495432968_!_110_!_news_military_!_若现代社会发生战争，普通家庭可做哪些简单准备？_!_
+6552394069639692547_!_110_!_news_military_!_叙利亚政府军从一处叛军据点缴获的武器里面有俄制的装备，你怎么看？_!_
+6552338603932910084_!_110_!_news_military_!_让40国军人对它都爱不释手，中国的这套军服到底有多大魅力_!_越南人民军,65式军服,年代,芳华,人民子弟兵
+6552313982080778504_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552441575052411149_!_110_!_news_military_!_你怎么看待以色列在一夜之间盗走伊朗上吨核文件这种行为？_!_
+6551715409354031624_!_110_!_news_military_!_妩媚女兵不让须眉：以色列女兵演训生活图照_!_义务兵役制,实战演习,以色列,兵役法,女兵
+6552464882082513166_!_110_!_news_military_!_战斗机被导弹锁定追踪后，战斗机能否发射空空导弹摧毁追踪导弹？_!_
+6552408921401721095_!_110_!_news_military_!_历史上的十大军事人才有谁？_!_
+6552274936860246286_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6550441849356550414_!_110_!_news_military_!_这次叙利亚战争后，叙利亚还能复兴吗？_!_
+6552417370059571464_!_110_!_news_military_!_为什么叙利亚40个反对派同时反对俄罗斯主导的和谈？_!_
+6552133095686930948_!_110_!_news_military_!_中国大西北飞出这一架战机了不得，空军实力陡增它立大功_!_隐身技术,陕飞,战斗机,预警机,机载雷达
+6552235700144570887_!_110_!_news_military_!_从军中娇娃到赛场魅影 她便是这样的女兵_!_赛车,女兵,赛车手,军人
+6552300684908167438_!_110_!_news_military_!_《刺激战场》换弹速度最慢的武器有哪些，性能如何？_!_
+6552336374815523342_!_110_!_news_military_!_印度军队打仗不戴头盔，而是戴头巾，是真的穷还是坚持信仰？_!_头盔,宗教信仰,戴头巾,头巾,印度军队
+6552453859426435342_!_110_!_news_military_!_为什么历史上没有一个国家能吞并中国？_!_
+6552240481537884419_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552344044410765571_!_112_!_news_travel_!_日落霞飞苍山远，求下句？_!_
+6552421666394210820_!_112_!_news_travel_!_茂陵博物一日游_!_茂陵博物,一日游,博物,茂陵
+6552496775318471171_!_112_!_news_travel_!_旅游风景电脑壁纸 适合旅游的自然风景_!_自然风景,电脑壁纸,风景
+6552128508959654408_!_112_!_news_travel_!_亚洲地理位置最好的国家_!_德干高原,印度,喜马拉雅山,恒河平原,孟加拉湾
+6552315559835336964_!_112_!_news_travel_!_四川游客张某带着六名女子在三亚沙滩烧烤被撒沙，这事你怎么看？_!_
+6552485616423010573_!_112_!_news_travel_!_藏南面积有多长多宽？_!_
+6552326138541113859_!_112_!_news_travel_!_微视频｜一分钟领略绵竹年画村”画景”_!_绵竹市孝德镇,绵竹年画,年画村,年画,绵竹年画村
+6552353840400695815_!_112_!_news_travel_!_东北雪乡景色美，网友称：真是一场斗智斗勇的旅行_!_东方ic,东方IC,景色,雪乡
+6552380093732225284_!_112_!_news_travel_!_勐景来，在拍完“爸爸去哪儿了3”后，如何做到持续吸引全国游客？_!_
+6552328246938042884_!_112_!_news_travel_!_旅行达人告诉你，旅行不管短途或是长途别忘了带这些，干净还卫生_!_衣架,马桶垫,报警器,高级显档次,旅行箱
+6552473513825927431_!_112_!_news_travel_!_一个人去爬山合适吗？_!_
+6552346786705441287_!_112_!_news_travel_!_发现贵州之美，尽情享受，难以忘怀_!_黄果树,天台山,屯堡,黄果树瀑布,平坝县
+6552363331502473479_!_112_!_news_travel_!_护照有效期不足6个月能去普吉岛吗？_!_
+6552392052749893902_!_112_!_news_travel_!_四川游客张某带着六名女子在三亚沙滩烧烤被撒沙，这事你怎么看？_!_
+6552250246221529607_!_112_!_news_travel_!_重庆彩色马赛克“梯田”吸引游人_!_重庆,马赛克,梯田
+6552261929702785549_!_112_!_news_travel_!_五月冰城变花海，看丁香花丛中的哈尔滨_!_丁香花,梨花,冰城,杏花,哈尔滨
+6552286499683959044_!_112_!_news_travel_!_想带二老去湖北游玩，非自驾，预计花费5000，有没有攻略路线推荐？_!_
+6552407884271976717_!_112_!_news_travel_!_户外直播跳舞的爽儿为什么出镜率那么高？_!_
+6552272024415240708_!_112_!_news_travel_!_「跟着大V去旅行」来到开滦，走进这座活着的博物馆_!_博物馆,开滦煤矿,开滦,蒸汽机车,中国工业
+6552464212268941581_!_113_!_news_world_!_哈萨克斯坦允许美国建立军事基地，如果哈彻底倒向美国，对俄罗斯究竟有多大威胁？_!_
+6552438918724190472_!_113_!_news_world_!_印度一男子不顾劝阻要和熊自拍，被咬死，大家怎么看？_!_
+6552397094101975299_!_113_!_news_world_!_美国建国不过两百年，为什么会成为世界强国？_!_
+6552267481996591623_!_113_!_news_world_!_警告！夏威夷火山正在向空中喷射熔岩和“极高水平的危险气体”_!_火山爆发,夏威夷,裂缝,熔岩,地震
+6552299188359528974_!_113_!_news_world_!_印度大叔来到中国后很震撼，在中国平常的事情在印度是不可思议的_!_印度人,印度,三峡大坝
+6552377486179566088_!_113_!_news_world_!_45岁男子一日三餐都是吃石头砖头，啃石头就像是吃着美味的鸡腿_!_迈克尔
+6552278400227082499_!_113_!_news_world_!_美国建国不过两百年，为什么会成为世界强国？_!_
+6552311506430591491_!_113_!_news_world_!_特朗普为挺拥枪嘲讽巴黎恐袭，法国怒怼“可耻”_!_总统,布雷迪,2017年枪支暴力伤亡统计,特朗普,民主党,枪击案
+6552376898788262408_!_113_!_news_world_!_日本拼外交，诚意不可少_!_自民党,安倍晋三,日本,吕耀东,朝日新闻,安倍内阁
+6552260360039039245_!_113_!_news_world_!_能够连接印度洋和太平洋的泰国克拉运河项目是否已经开工，一旦建成对新加坡有影响吗？_!_
+6552233452341559565_!_113_!_news_world_!_如果证实反政府武装使用化学武器，美英法会被追责吗？_!_
+6552400172599476493_!_113_!_news_world_!_美国建国不过两百年，为什么会成为世界强国？_!_
+6552313888182894851_!_113_!_news_world_!_美国法院判伊朗赔付9·11遇难者家属逾60亿美元，你怎么看？_!_
+6550517223964803336_!_113_!_news_world_!_法国劝说奏效，伊朗为打消顾虑愿加入对话，特朗普还会退出伊核协议吗？_!_
+6550828176661020932_!_113_!_news_world_!_梅德韦杰夫对普京来说有多重要？_!_
+6552396908315279880_!_113_!_news_world_!_只因朋友称《绝地求生》更好，“堡垒”男子将其刺伤_!_绝地求生,佛罗里达州,Andrew,David,美国
+6552336742647595271_!_113_!_news_world_!_作为俄罗斯总统，普京上台以来有哪些失误？_!_
+6552469217210269960_!_113_!_news_world_!_从目前来看，特朗普是一个好总统吗？_!_
+6552153869697155335_!_113_!_news_world_!_如果没有普京的出现，世界格局将会变成什么样？_!_
+6552486135359078660_!_113_!_news_world_!_特朗普为什么可以随意废除此前美国政府达成的各种协议？_!_
+6491869702602621198_!_114_!_stock_!_为什么有的股票在大盘跌时它涨，在大盘涨时它反跌？_!_中国银行,同方股份,两桶油,江南嘉捷,上证指数,股票,上海医药,工商银行,大盘
+6550942691302048259_!_114_!_stock_!_中国限产，全球缺货，这家有机硅龙头必须关注起来了！_!_草甘膦,DMC,副产物,有机硅,新安股份
+6531148042983178509_!_114_!_stock_!_大疆的概念股有哪些？_!_
+6446238642045190413_!_114_!_stock_!_高位十字星是好是坏？_!_股价,个股,收盘价,开盘价,高位十字星
+6547887045119836429_!_114_!_stock_!_一支股票卖盘一直挂豹子单，有何用意？_!_
+6477380983082451214_!_114_!_stock_!_底部放量是什么意思？_!_公司基本面,宏达股份,限售股,江苏银行,华联综超
+6481098178765521166_!_114_!_stock_!_股票放量下跌是什么意思？放量下跌意味着什么？_!_成交量,股票,股票软件,放量,投资者
+6487332401843273998_!_114_!_stock_!_什么情况下我们需要毫不犹豫的卖出股票？_!_股票,小阴线
+6529752296744550664_!_114_!_stock_!_股票在涨停后，半小时又下去了，再过一会又涨停1个小时，随后又下去了，这说明什么？_!_
+6493440590351433998_!_114_!_stock_!_有些股票前一天涨停，后一天就跌停，都是主力所为吗？主力看样子亏了呀？_!_主力,无能为力,跌停,涨停板,獐子岛
+6488447166112071950_!_114_!_stock_!_怎么样判断一只股票是否有庄家锁筹？_!_均线,股票,支股票,投票器,基本面
+6520037319762772227_!_114_!_stock_!_连续跌停的股票是谁在买入？_!_成交量,千山药机,保千里,散户,投资者
+6431870310357664001_!_114_!_stock_!_如何做T+0！一招轻松解套？_!_解套,辅助线,投资者,股票,金字塔式
+6491752508132688142_!_114_!_stock_!_大盘跳水的时候个股跌停板一直没封死，庄家的这种行为是酝酿的什么阴谋？_!_茅台酒,技术分析,大盘,股票,股市
+6493788508598042893_!_114_!_stock_!_次新股有望开启反弹行情吗？_!_kdj指标,次新股,新股,堰塞湖,投资者
+6528576283499036941_!_114_!_stock_!_芜湖有哪些上市公司？_!_
+6534628622483849479_!_114_!_stock_!_股票头天封涨停，第二天低开是怎么回事？_!_
+6552252565776499213_!_115_!_news_agriculture_!_鲜蒜、蒜薹今年恐怕“狠”不起来 多地出现免费送_!_鲜蒜,上货量,蒜薹
+6551606480724820228_!_115_!_news_agriculture_!_咖啡兰的养殖要点是什么，和其他品种有什么区别？_!_
+6552465334081683719_!_115_!_news_agriculture_!_要知询农业专家，象怎样的高温天气，大地的西红柿与豆角是否天天浇水？_!_
+6552294857350578702_!_115_!_news_agriculture_!_农村盖房子的物料，虽然很廉价，它的药用价值，你们都不知道_!_致病微生物,抗生素,缓凝剂,石膏,单斜晶系
+6552392600622465539_!_115_!_news_agriculture_!_2018年农村房产证已陆续发放，但这些人没有机会领取！_!_建房,宅基地,土地改革,产权证,农村房产证
+6552154540022432269_!_115_!_news_agriculture_!_农村新生代模特_!_模特,新生代,农村
+6552483708689973764_!_115_!_news_agriculture_!_2018宅基地新动作：满足这4个条件的农民，可进行分户_!_
+6552397961869918477_!_115_!_news_agriculture_!_有50万的资金买车还是买地？_!_
+6552221575469334787_!_115_!_news_agriculture_!_玉米“多穗”现象形成的原因有哪些？该怎么预防？_!_
+6552253580173115907_!_115_!_news_agriculture_!_农村承包地确权进入收尾阶段 确权面积已达11.59亿亩_!_全国农村承包地,承包地,土地权属,农村,承包地确权登记颁证
+6552234153524331015_!_115_!_news_agriculture_!_来广元考察交流的巴中市妇联，看看她们都去了哪？_!_互助会,广元,企业家,妇女儿童,广元市巾帼建功
+6552415602848301582_!_115_!_news_agriculture_!_家里有空地，不如种几棵果树，随摘随吃，吃都吃不完_!_果实,红如玛瑙,李子,树苗,桂樱亚属
+6552378317746471427_!_115_!_news_agriculture_!_关于南半球的六道题你会几个？测一测吧？_!_Winery,阿根廷,Toro,Rothschild,赤霞珠
+6552466309865538056_!_115_!_news_agriculture_!_鸭稻算什么，看看我们龙江的“天鹅稻”！_!_富锦市,有机质,贝加尔湖,黑龙江,建三江管理局
+6552403437995163918_!_115_!_news_agriculture_!_想用十五万来创业，有哪些适合做的项目？_!_
+6552131650216526344_!_115_!_news_agriculture_!_大蒜产地蒜薹大面积滞销，农民辛劳一天，蒜薹还是要扔丢_!_蒜薹,大蒜,东方ic,东方IC
+6552397931356357123_!_115_!_news_agriculture_!_农村大叔野外抓螃蟹，却收获了这么多的小石头！_!_农村,螃蟹,大叔
+6552445602754462211_!_115_!_news_agriculture_!_温室西红柿怎么种植？西红柿现代温室种植_!_基质,珍珠岩,烟雾剂,可湿性粉剂,西红柿
+6552368060429238798_!_115_!_news_agriculture_!_撑着小船来娶你！悠悠隆庆，用500年岁月诉说水乡故事_!_罗南,陈氏宗祠,古官道,隆庆村,隆庆
+6552324432327606791_!_115_!_news_agriculture_!_我国首座全潜式深海渔场5月4日在青岛交付使用_!_深海渔场,全潜式,八边形,三文鱼,养殖
+6552404869309792515_!_116_!_news_game_!_kpl明星赛最大的败笔是不是替补坐了两场冷板凳？_!_
+6552483188243956231_!_116_!_news_game_!_绝地求生：近日爆料即将新增北极地图，那些吃鸡神器直接废了！_!_新增北极地图,吉利服,制作人,地图,绝地求生
+6552396006221152516_!_116_!_news_game_!_为什么国产手游做不出好游戏？_!_
+6552420119983685902_!_116_!_news_game_!_LOL中哪个英雄最克劫？_!_
+6552089978770293251_!_116_!_news_game_!_索泰杯第二日QM吃两鸡引发争议？虎扑大神扒出QM违规行为？_!_贺岁杯,QM,战队,索泰杯,QM战队
+6552133414735053315_!_116_!_news_game_!_棋牌游戏运营：2018上半年棋牌游戏总结及展望_!_地方棋牌,棋牌游戏市场规模,房卡模式,游戏平台,棋牌行业
+6552362717309567235_!_116_!_news_game_!_现在王者荣耀都禁哪些英雄？_!_
+6509445913239355655_!_116_!_news_game_!_“CF”都出了英雄级，为什么还要出“王者系列”？_!_穿越火线,王者,英雄级武器,王者系列,玩家
+6552415610976862724_!_116_!_news_game_!_DNF：大家自嘲一下自己最缺哪件诗史_!_游戏,诗史套,dnf,缺腰带,诗史
+6552445536111165710_!_116_!_news_game_!_有什么唯美剧情的手机单机游戏？_!_
+6552345991649952270_!_116_!_news_game_!_免费畅玩Dota2所有饰品！DOTA2 MOD软件安利_!_自由度,懒人包,饰品,大部分,dota2
+6552440079824978436_!_116_!_news_game_!_携NBA完美还原球星生活最强NBA揭秘潮服、王朝两大新模式_!_最强NBA,雷阿伦,潮服,总决赛,中南大学
+6552476398626275844_!_116_!_news_game_!_王者荣耀：当队友选这三个奇葩英雄打野的时候，我就知道此局必输_!_打野英雄,打野,姜子牙,荣耀第一线
+6552083347386925576_!_116_!_news_game_!_王者荣耀：反野最厉害的四个英雄，最后一个比东皇太一还恶心！_!_钟馗,反野,东皇太一
+6552481588993589518_!_116_!_news_game_!_三国杀里有哪些嘲讽度很高的武将？_!_
+6552342754154775043_!_116_!_news_game_!_区块链游戏哈希世界新增南极土地玩法，可领比特币_!_哈西,哈希世界,南极土地,自动发注册,南极大陆
+6552445691136836100_!_116_!_news_game_!_绝地求生黄金大奖赛TPP模式奖金36万、FPP奖金10万 奖金差距好大_!_大奖赛,第三人称,FPP,TPP
+6552452952596939016_!_116_!_news_game_!_有什么好玩的天文游戏？_!_
+6552436208524329475_!_116_!_news_game_!_Uzi直播首秀人气火爆弹幕刷到黑屏，黄金段位拿五杀称太简单！_!_uzi,MSI,Uzi,艾欧尼亚,LPL
+6552327929114657027_!_116_!_news_game_!_王者荣耀中吕布出闪电匕首、纯净苍穹或者无尽之刃哪个性价比或者效果好一点？_!_
+6552396255467667720_!_101_!_news_culture_!_"皇"和“王”哪个地位更高？_!_
+6552307476732051716_!_101_!_news_culture_!_上联：江中鱼东游西藏，下联怎么对？_!_
+6552340726946988296_!_101_!_news_culture_!_上联：若不撇开终是苦，如何对下联？_!_
+6552299286422356227_!_101_!_news_culture_!_出：窗含春夏秋冬景，下联怎么对？_!_
+6552301997410746638_!_101_!_news_culture_!_上联。山西山东太行山。下联？_!_
+6552294124047827213_!_101_!_news_culture_!_为什么好多人都说儒家思想祸害中国？_!_
+6552286376518222350_!_101_!_news_culture_!_曹雪芹墓地石碑和玉器被村民发现，上交给国家，被奖励一面锦旗_!_块墓碑,曹雪芹,石碑,玉器
+6552371576543641864_!_101_!_news_culture_!_上联：野花无主为谁开，如何对下联？_!_
+6552333690825867789_!_101_!_news_culture_!_中国历史上六大顶级国宝级文物，传国玉玺才排第三，第一无可争议_!_传国玉玺,永乐大典,史记·天宫书,丧乱帖,丧乱贴,文物,敦煌遗书,金刚般若波罗蜜经,王羲之,大英百科全书,莫高窟
+6552471931298251272_!_101_!_news_culture_!_纪念伟人，马克思诞辰200周年！_!_私有制,马克思主义,国际共产主义运动,马克思,共产党宣言,共产主义
+6552240748727632397_!_101_!_news_culture_!_难得一见的素描作品，书画爱好者的福音！_!_书画,素描作品
+6552474352665755917_!_101_!_news_culture_!_上联：拜托流云捎往事，如何对下联？_!_
+6552142305137000711_!_101_!_news_culture_!_为何张辽和高顺会被陈登所骗？_!_
+6552407325947199752_!_101_!_news_culture_!_搞艺术的人有什么共性？_!_
+6552389460921680391_!_101_!_news_culture_!_脍炙人口的醉翁亭原来是这个样子，这里的山水的确很美_!_醉翁亭,意在亭,醉翁亭记,醉翁吟,滁州,琅琊山,欧阳修
+6552447148435177998_!_101_!_news_culture_!_千年德国小镇居民自称为“中国人”，中文是其官方语言之一_!_天安门广场,博物馆,官方语言,德国小镇
+6552348386157134339_!_101_!_news_culture_!_一组清代光绪年间的壁画_!_函谷关,壁画,光绪年间
+6552409239053140488_!_101_!_news_culture_!_以玉定情，始结同心_!_香香公主,陈家洛,霍青桐,古诗笺
+6552299663423176968_!_101_!_news_culture_!_中国十八世纪时，都有哪些著名的诗人？_!_
+6552413454714536196_!_101_!_news_culture_!_上联：梨开一树雪。下联是什么？_!_
+6552849380540416515_!_102_!_news_entertainment_!_同样懒精上身群众却只喷张杰不喷他，这是双标吗？_!_何炅,刘宪华,池子,李诞,情深深雨蒙蒙,张绍刚,彭昱畅,张杰,干活,黄磊
+6552378214478512387_!_102_!_news_entertainment_!_如果把范冰冰嫁给你，你敢娶吗？_!_
+6552479146792976900_!_102_!_news_entertainment_!_一天烧近290万元，现今，它到底有多重要_!_复仇者联盟3 ：无限战争,片酬,复仇者联盟,复联3,复仇者联盟3：无限战争,电影,复联4,视觉特效
+6552381535213847053_!_102_!_news_entertainment_!_张若昀和胡一天都很帅气，他们不撞脸，却撞痣了_!_金喜善,刘亦菲,张若昀,全智贤,韩佳人
+6552377227537809928_!_102_!_news_entertainment_!_邓超母校这么多？现场瞬间变身国际学霸超 却遭陈赫无情补刀_!_跑男,老邓头,哈佛大学,龙舟比赛,陈赫,邓超
+6552380377560777223_!_102_!_news_entertainment_!_郭敬明现身机场，粉丝跪地拍照？网友：这样做也太夸张了吧_!_郭敬明,摄像师,粉丝,机场
+6552371750187827459_!_102_!_news_entertainment_!_有哪些明星明明长相很普通，却总演高富帅？_!_
+6552336879226716686_!_102_!_news_entertainment_!_马栏山 马栏镇 马栏村 马栏坡的马小姐获两项吉尼斯世界纪录称号_!_吉尼斯世界纪录,马栏坡,马栏镇,快乐大本营,马栏山,谢娜
+6552341424665264648_!_102_!_news_entertainment_!_好莱坞7位惊世美男也禁不住岁月的琢磨_!_茜茜公主,教父1,教父2,盗梦空间,教父3,荒野猎人,加勒比海盗,泰坦尼克号,黑客帝国,剪刀手爱德华,碟中谍,佐罗
+6552376917822013966_!_102_!_news_entertainment_!_Angelababy又接新剧，这次男主是邓伦和朱一龙，你期待？_!_新剧,朱一龙,Angelababy,angelababy
+6552362801082401283_!_102_!_news_entertainment_!_秦岚拍婚纱，无比美！网友：今天，你要嫁给我？_!_秦岚,嫁给我,婚纱
+6552319871793758723_!_102_!_news_entertainment_!_《甄嬛传》中安陵容这么乖巧、听话，皇帝还对安陵容如此绝情？_!_安陵容,皇帝,甄嬛,甄嬛传,元皇后
+6552451239223755267_!_102_!_news_entertainment_!_25届北京大学生电影节闭幕式，芳华演员钟楚曦、苗苗获最佳新人奖_!_钟楚曦,最佳新人奖,闭幕式,北京大学生电影节,演员
+6552357408776126990_!_102_!_news_entertainment_!_曹格的老婆在微博晒出了一组带儿女的合照 姐姐的大长腿令人羡慕_!_爸爸去哪儿,Grace,曹格,姐姐,我是歌手,萌可爱,Joe
+6552295488261980679_!_102_!_news_entertainment_!_什么时候开始都不晚！宗庆后亮相《朗读者》第二季，他是这样读给年轻新一代的_!_片儿川,宗庆后,娃哈哈,季羡林,平平常常,八十抒怀,朗读者
+6552407641648267780_!_102_!_news_entertainment_!_女星还债4亿，拿1亿现金买房，王珂追问：刘涛，你还有多少钱_!_刘涛,王珂,不离不弃,女星,演艺圈
+6552111599363031565_!_102_!_news_entertainment_!_金晨出席活动现场表演 网友：对，这大长腿就应该扛在肩膀上！_!_金晨,长腿,金晨出席活动现场表演,出席活动现场表演网友
+6552290381810106893_!_102_!_news_entertainment_!_低调多年，赵丽颖哥哥终曝光，长得帅零绯闻，身家不比赵丽颖少！_!_追鱼传奇,钻石王老五,偏偏爱上你,赵丽颖,乘风破浪,丁子峻,跪族篇,陆贞传奇,韩雪,Angelbaby
+6552130645336785165_!_102_!_news_entertainment_!_为什么扎扎帕楚里亚被网友戏称为“闪垫侠”？_!_
+6552337351002030606_!_102_!_news_entertainment_!_生活中的太子妃张天爱 网友：比做太子妃美艳_!_父子雄兵,凌织羽,鲛珠传,落樱,张天爱,太子妃升职记,武动乾坤,奇幻电影,建军大业,妖猫传,二炮手,王牌对王牌3,文荣奖,张芃芃
+6552159996241510659_!_103_!_news_sports_!_19年亚洲杯国足所在的C组算是上上签吗？_!_
+6552409756328264205_!_103_!_news_sports_!_安德森数据依旧犀利！对比之后并没有变化，德安东尼在下一盘大棋_!_安德森,常规赛,季后赛,巴莫特,爵士队
+6552313655503880456_!_103_!_news_sports_!_勒布朗已是美国各大媒体公认的历史第二人，他与乔丹的差距究竟还有多大？_!_
+6552329024117408263_!_103_!_news_sports_!_骑士战胜猛龙拿到赛点，詹姆斯此项数据NBA第一，超过乔丹_!_骑士,季后赛,猛龙,詹姆斯,NBA
+6552317604382376462_!_103_!_news_sports_!_凯西谈德罗赞：投不好也可以做其它事，比如防守_!_伊巴卡,詹姆斯,凯西,德罗赞
+6552265019709456904_!_103_!_news_sports_!_24+5+4，第一新秀换人！20岁天才刷爆72年历史，小詹皇被他打爆_!_凯尔特人,NBA,塔图姆,季后赛,费城76人
+6552412711362232840_!_103_!_news_sports_!_人和逼平鲁能，京城狂涨粉，北京国安下轮有望登顶！_!_广州富力,中超联赛,山东鲁能,山东鲁能泰山,北京国安
+6552288738657960195_!_103_!_news_sports_!_瓦基弗银行3:2逆转意大利冠军，晋级欧冠决赛，距离全满贯又近了一步，怎么看？_!_
+6552406355297173773_!_103_!_news_sports_!_中超第九轮，山东鲁能0:0战平北京人和，错失登顶良机，怎么评价本场比赛？_!_
+6552467414066397454_!_103_!_news_sports_!_为什么提起火箭队，很多球迷就会想起奥拉朱旺？_!_
+6552312669032939783_!_103_!_news_sports_!_世乒赛夺冠后，刘诗雯调侃自己在职业生涯末端，两年后奥运会还能见到小枣吗？_!_
+6552291442427953677_!_103_!_news_sports_!_2018年俄罗斯世界杯12座球馆巡礼——伏尔加格勒体育场_!_体育场,俄罗斯,伏尔加河,伏尔加格勒,俄罗斯杯
+6552304132349231373_!_103_!_news_sports_!_詹皇绝杀乐福21+16，骑士险胜猛龙总比分3-0，你怎么看？_!_
+6552273611615371780_!_103_!_news_sports_!_联赛8轮不败+斩落领头羊 国安搅乱中超争冠格局_!_中超,施密特,鲁能,韦世豪,埃尔克森
+6552313622704423176_!_103_!_news_sports_!_2019亚洲杯中国国足到底怎样做才能“夺冠”？_!_
+6552463488202048014_!_103_!_news_sports_!_一水二浪三死神 浓眉做花迪做叶 横批？_!_死神,杜兰特,浓眉,金州勇士队,霍乐迪
+6552292746734862855_!_103_!_news_sports_!_杰森-特里：塔特姆打出了超过他年龄的极佳表现_!_杰森-,NBA,杰森,塔特姆,凯尔特人
+6552455225951650307_!_103_!_news_sports_!_多点开花，国际米兰客场大胜乌迪内斯_!_达尼洛,拉菲尼亚,佩里西奇,乌迪内斯,伊卡尔迪
+6552430545446371843_!_103_!_news_sports_!_生猛！这个女人当4年马刺助教，现在又要去应聘主教练！_!_雄鹿队,俄罗斯,主教练,贝基·哈蒙,马刺
+6552478006881813000_!_103_!_news_sports_!_离被横扫不远？猛龙系列赛6困扰难解决：1优势骑士轻松破除_!_猛龙,德罗赞,系列赛,骑士,瓦兰丘纳斯
+6552374236030173700_!_104_!_news_finance_!_烟台企业参加广交会入账1.23亿美元_!_广交会,烟台市
+6552183305750446600_!_104_!_news_finance_!_「上币公告」ABCC上线DASH、OMG_!_DASH,ETH,达世币,BTC,ABCC
+6552457929671311885_!_104_!_news_finance_!_深市今日停复牌一览(5月7日）_!_股份2018-05-0709,30取消停牌重大,停牌
+6552283541407793671_!_104_!_news_finance_!_广交会到会采购商与成交额实现双增长_!_广交会,采购商,春交会
+6552022695070925326_!_104_!_news_finance_!_职业操盘手不可外传的成交量买卖精髓，字字珠玑！_!_主力,量能,沙河股份,股价,投资者
+6552310030106886659_!_104_!_news_finance_!_活动预告l 资管新规解读暨债市投资策略会_!_天府新区,资管新规,对冲基金,瑞银,国投瑞银基金
+6552401859213000963_!_104_!_news_finance_!_余额宝升级后，有两只新的货币基金，分别是博时基金和中欧基金管理，对用户有什么风险吗？_!_
+6552274692940497412_!_104_!_news_finance_!_创业者应该掌握的创业理念和方法_!_创业者
+6552187676118745614_!_104_!_news_finance_!_陈柯枫：5.6黄金将继续反弹，下周操作策略_!_伊朗核协议,整数关口,黄金,美国,避险情绪
+6552452707783803399_!_104_!_news_finance_!_“擒龙术”核心——抓龙头让你不错过题材或者板块龙头的任何机会_!_操作方式,方大炭素,龙头个股,海特高新,安妮股份,宝泰隆,知识产权,操作
+6552242528677003779_!_104_!_news_finance_!_信用卡逾期，已还本金10万，银行还要催收15万逾期利息怎么办？_!_逾期利息,信用卡,银行,日利率,利息
+6552298859064721934_!_104_!_news_finance_!_GuruFocus创始人：股神接班人是今年最大亮点_!_巴菲特,Ajit Jain,伯克希尔,Abel,接班人
+6552335213878313223_!_104_!_news_finance_!_普通外汇投资者该如何炒外汇？_!_
+6552464466150162952_!_104_!_news_finance_!_5月6日股市内参_!_药明康德,股神巴菲特,亚马逊,内布拉斯加州,上证指数,沙河股份,梦佳
+6552469167885254919_!_104_!_news_finance_!_长线投资一般投资什么股票？_!_
+6552365156284760583_!_104_!_news_finance_!_山西首富是个初中生，第二名和第三名的财富加起来也没首富一半多_!_杨建新,百圆裤业,中国人民大学,清徐县,经济学
+6552301338103906824_!_104_!_news_finance_!_是走还是留？15.60元的成本，只怕是无影脚的票，第二天高开出货_!_财政局,新昌县,事不过三,力科技,金融
+6552744415880806915_!_104_!_news_finance_!_被大多数人忽略的买房好处，我猜你没有想到！_!_买房,郑州,房价,炒外汇,银行
+6552170138244743438_!_104_!_news_finance_!_准备卖大米，面粉，油，应该在哪里进货？_!_
+6552383398747308302_!_104_!_news_finance_!_“国六”即将实施，常用的“国三”“国四”会怎样处理呢？_!_
+6552277796377329928_!_106_!_news_house_!_山西运城的房价近期会有大的上涨吗？_!_
+6552315323750547725_!_106_!_news_house_!_子女和父母在同一个户口下，如何将子女名下的房产过户给父母？_!_
+6552316131120513288_!_106_!_news_house_!_房价会不会再次上涨？_!_
+6552376128122650884_!_106_!_news_house_!_现在在县城买房合适吗？_!_
+6552448004974969348_!_106_!_news_house_!_多个好盘入市！2406套优质房源今起登记！_!_优先选房,货币化,户型,优先摇号,房源
+6552414447514681614_!_106_!_news_house_!_一个城市的发展靠的是什么？_!_
+6552451353103302925_!_106_!_news_house_!_为什么现在还有人在等着房子降价再买？_!_
+6552316481919516936_!_106_!_news_house_!_结婚十年公婆想全款买一套房子写他儿子的名字，生活中如果发生意外，妻子能分这套房吗？_!_
+6552253461335900419_!_106_!_news_house_!_我是天水人想去西安买房定居，我的选择正确吗？_!_
+6552281574245663239_!_106_!_news_house_!_买房最高补贴200万，50万安家补助！安徽这个市PK合肥人才新政！_!_芜湖,给予,补助,芜湖市,战略性新兴产业,安家费
+6552305714935628296_!_106_!_news_house_!_强势围观！莱芜楼市迎来“大洗牌” 这些新盘即将入市 热度爆表！_!_中国地产,碧桂园,莱芜,新项目,金三银四
+6552322994490835203_!_106_!_news_house_!_成都本地人怎么看待成都的房价？_!_
+6552383464606269965_!_106_!_news_house_!_「司法拍卖：理想小镇」双拼别墅 带车位/车库 南北朝向 可贷款_!_建筑面积,福德站,昆明市,幢101号别墅房屋,大商汇
+6552337214938808584_!_106_!_news_house_!_相比起大庆的工资水平，你觉得大庆的房价高吗？_!_
+6552352198880133646_!_106_!_news_house_!_深圳房价连降19个月，为什么买房者还嫌不给力？_!_深圳市,降价,楼市,深圳,房价
+6552165001036038414_!_106_!_news_house_!_西安的房价还会涨吗？_!_
+6552266013163913486_!_106_!_news_house_!_炒房是不是已经向小县城转移了？_!_
+6552324329537798407_!_106_!_news_house_!_一旦房子价格暴跌，我们会面对怎样的场景？_!_
+6552260668991471876_!_106_!_news_house_!_为什么越来越多的安顺年轻人到贵阳买房？_!_
+6552340144026812686_!_106_!_news_house_!_建房屋第一层最高可以多少米？_!_
+6552139303667892739_!_107_!_news_car_!_大家都说日本车不安全，你还会买日本车吗？_!_昂克赛拉,马自达,日本车,国产车,日本
+6552418521509265934_!_107_!_news_car_!_最美大众，CC强势回归，网友：更加漂亮了_!_大众CC,CC,汽油发动机,大众cc
+6552181354593780237_!_107_!_news_car_!_夏日高温来，车内烫的如火烧身！装冰丝坐垫，开车出行凉爽又舒适_!_汽车坐垫,四季通用,科鲁兹,冰丝,汽车座垫,坐垫
+6552268969774940430_!_107_!_news_car_!_为什么有些开好车的女生很漂亮？_!_
+6552317675391943182_!_107_!_news_car_!_夏天来了，去体验一把住房车呗！_!_房车露营,房车,信用卡,桌上足球,乒乓球
+6552496028732359175_!_107_!_news_car_!_6.4入手的大众高尔夫，全靠老爷子一人入手，千万不能得罪老人家_!_二手车,大众高尔夫,老爷子
+6552244748118131203_!_107_!_news_car_!_为什么爱车 空调不制冷？大多数车主不清楚，搞懂后车内清凉一夏_!_车睡神器,空调清洗剂,汽车轮胎,汽车空调,清洗剂
+6552353743843623438_!_107_!_news_car_!_安全倒车9不要_!_天津
+6552443540461322759_!_107_!_news_car_!_被誉为日本劳斯莱斯 丰田顶级豪车 它才是政商大佬身份的象征_!_丰田,V12,丰田世纪,w140,大板车,烟灰缸
+6552360908482412803_!_107_!_news_car_!_转型中的长安汽车，推高端品牌，7年内能达到600万的销售规模吗？_!_
+6552350134892495364_!_107_!_news_car_!_小米作为手机行业的佼佼者，他在汽车行业何去何从_!_国产汽车,特斯拉,新能源汽车,汽车行业,小米
+6552404779811734020_!_107_!_news_car_!_停车先拉手刹还是先挂P档？老司机3句话让你茅塞顿开，太精辟_!_自动挡,钩子,拉手刹,排档杆,变速器
+6552281868497060359_!_107_!_news_car_!_纱窗坏了，别去维修 18年特流行这样装，防蚊防虫又保护隐私_!_充气床,磨砂玻璃,掏耳朵,胶静电
+6552257676884050439_!_107_!_news_car_!_本来不打算买Giulia，但是听说降到30.08万了_!_万元3.00,经销商,Giulia,减速带,R19
+6552296823027925517_!_107_!_news_car_!_你的转发会帮助很多人！_!_长春市,网约车,行车证
+6552439660709151240_!_107_!_news_car_!_在北京国际车展领克展台……仿佛去了一趟游乐场？_!_新能源,PHEV,欧洲市场,克02,吉利汽车,CEVT,CMA,混合动力
+6552353263784559117_!_107_!_news_car_!_做到这几点，发动机用20年都不是问题！幸亏我知道得早_!_发动机,曲轴箱,热车,机械能,积碳
+6552308154145702413_!_107_!_news_car_!_汽车销售小姐最怕消费者来这3招，这些销售内幕让人心寒！_!_销售员,汽车销售,买车,4s店
+6552344843513758216_!_107_!_news_car_!_雪铁龙在华销量低迷却喊出“因你不同”_!_爱丽舍,X-RV,云逸,中国市场,IX25,雪铁龙,C4L,东风雪铁龙,雪铁龙C4L
+6552352000644743437_!_107_!_news_car_!_同一个品牌的高配汽车和低配汽车的发动机一样吗？_!_
+6552480119196221966_!_108_!_news_edu_!_家长们看过来，南充清泉坝将建2所学校，孩子上学不用愁_!_南充,清泉坝,华丽转身,菜园子,顺庆区
+6552332706087174408_!_108_!_news_edu_!_孩子学习成绩全班倒数，为什么老师不能抽时间为孩子补补课，老师不应该帮孩子提高成绩吗？_!_
+6552307452954542344_!_108_!_news_edu_!_考试，县级事业单位和公务员，应该选哪个？_!_
+6552128674819211528_!_108_!_news_edu_!_怎样提高阅读能力？_!_
+6552437046760178183_!_108_!_news_edu_!_县城地区小学教研片英语绘本课堂教学大赛在肥东六中顺利举行_!_教师,县城,小学英语,县城地区,绘本
+6552411559723794692_!_108_!_news_edu_!_如何有效学习？_!_
+6552425151340741124_!_108_!_news_edu_!_酷炫！成都一高校情侣毕业返校拍另类婚纱照_!_婚纱照,成都,攀枝花,成都理工大学
+6552427614156358151_!_108_!_news_edu_!_教育那些事：高考后，优势专业与学校的选择（二）_!_电气自动化,电子信息工程,类专业,电子信息产业,电子信息
+6552353874932400392_!_108_!_news_edu_!_你认为班主任，老师，学生三者之间的关系是什么？_!_
+6552336240476160264_!_108_!_news_edu_!_为什么山东大学会逐渐没落？_!_
+6552457293244399879_!_108_!_news_edu_!_攀枝花哪些中学教学质量好？_!_
+6552394978973188365_!_108_!_news_edu_!_高考分数在二本中段线，想报临床医学专业，有希望吗？_!_
+6552431945219834372_!_108_!_news_edu_!_河北高校毕业生注意！你关心的就业创业政策来啦，看仔细→_!_就业服务,微型企业,高校毕业生,社会保险,助学贷款
+6552308584842002701_!_108_!_news_edu_!_上联：窗前清风拂月，有哪些合适的下联？_!_
+6552369133067960846_!_108_!_news_edu_!_大学学艺术是在浪费钱？将来挣的钱要比现在花的多_!_艺术,艺术市场
+6552385744189522435_!_108_!_news_edu_!_浙江财经大学_!_会计学,中外合作,金融学,奖学金,高级职称
+6552458248950120712_!_108_!_news_edu_!_怎么理解网上对这句「I like you, but just like you」 的翻译？_!_
+6552412407409410573_!_108_!_news_edu_!_原来暑假可以这样玩，家门口的异国文化盛宴！_!_关东煮,嘉年华,喜乐嘉年华,天方夜谭,马来西亚
+6552434977550631432_!_109_!_news_tech_!_一夜之间，芝麻分更新，7亿支付宝用户沸腾，你芝麻分涨了多少？_!_此时此刻,互联网金融,实名认证,支付宝,芝麻分
+6552358736696967694_!_109_!_news_tech_!_码商的福音！1688把货源搬进支付宝二维码 进货收钱一“码”事_!_小微企业,延期付款,数字化,支付宝,二维码
+6552286856447263246_!_109_!_news_tech_!_传小米阿里舜宇等八家公司列首批CDR名单_!_百度,独角兽,首次公开募股,CDR意见,CDR,小米
+6552287158651060493_!_109_!_news_tech_!_如何查找静电？_!_
+6552332417779106318_!_109_!_news_tech_!_区块链极简史2：比特币的转账机制 CCR炒币机器人_!_比特币,转账,矿工记账确认,手续费
+6552420737083245059_!_109_!_news_tech_!_工信部：我国将加快制定服装大规模定制相关标准_!_孙瑞哲,曹学军,大规模定制,中国服装协会,韩永生
+6552273484951585284_!_109_!_news_tech_!_自动驾驶、医疗、能源……一季度融资最多的人工智能公司_!_人工智能,机器人,Tempus,自动驾驶,Capital
+6552262852969758989_!_109_!_news_tech_!_几乎每一个银行工作人员都劝说顾客，不要把钱存入余额宝，对此你怎么看？_!_
+6552319796417921284_!_109_!_news_tech_!_用微信做淘宝客怎么样才能精准吸粉？_!_
+6552265044569096711_!_109_!_news_tech_!_雷军当年提出飞猪理论的时候，一定没想过会这样害人不浅！_!_飞起来,雷军,互联网,猪理论,人工智能
+6552475490827895300_!_109_!_news_tech_!_黑科技领路人！三星S10再曝：100%真全面屏稳了！_!_三星S10,屏手机,小米MIX,三星,Synaptics
+6552243666394546436_!_109_!_news_tech_!_经济学家吴敬琏为什么反对“不惜代价发展芯片产业”？_!_
+6552245081091342595_!_109_!_news_tech_!_小米会布局除cpu以外更多的芯片吗？_!_
+6552389188782653955_!_109_!_news_tech_!_恶搞漫画：金融危机对生活的影响_!_金融危机
+6552258243723264519_!_109_!_news_tech_!_查理芒格谈人工智能：炒作过多 还不能大规模取代人类_!_芒格,人工智能,机器智能,查理芒格
+6552402238512300291_!_109_!_news_tech_!_几乎每一个银行工作人员都劝说顾客，不要把钱存入余额宝，对此你怎么看？_!_
+6552310121278472712_!_109_!_news_tech_!_科普：使用微信付钱，这两个安全功能一定要打开，不然后悔！_!_数字证书,支付宝,微信
+6552396325411881486_!_109_!_news_tech_!_阿里旗下支付宝有梦想有情怀，码商成长计划承诺助力小微企业成长_!_支付宝,阿里,互联网,小企业,阿里巴巴
+6552430197126201603_!_109_!_news_tech_!_阿里云在全球处于一个什么地位？_!_
+6552477261017121027_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552229254107496707_!_110_!_news_military_!_如何看待美国将长时间停止对"白头盔"组织的资金供应？_!_
+6552397007984525571_!_110_!_news_military_!_中国辽宁航母战斗群的战斗力如何？_!_
+6552134542650507523_!_110_!_news_military_!_为什么西方国家不直接空袭叙利亚总统府，是怕吗？_!_
+6552384749879427598_!_110_!_news_military_!_美网络司令部升级 日裔陆军上将中曾根就任新司令_!_中曾根,马里兰州,司令部,网络司令部,陆军上将
+6552256872244576519_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552273573241684483_!_110_!_news_military_!_世界高超音速武器哪家强？中俄领先 美国已落伍_!_激光武器,高超音速武器,歼击机,俄罗斯,雪上摩托车
+6552432068331045134_!_110_!_news_military_!_红海行动中，队长他们解救人质后为什么不直接偷坦克离开，而要和恐怖分子爆发冲突？_!_
+6552392438671999246_!_110_!_news_military_!_二战时候，如果美国不参战，现在会是什么样？_!_
+6552269132195168516_!_110_!_news_military_!_如果现在一架最先进的战斗机穿越至秦朝坠毁，你觉得秦朝人能逆向出什么黑科技？_!_
+6550962683154268419_!_110_!_news_military_!_明知木质甲板易燃防护性差，为何二战时日本航母偏要用木质甲板？_!_
+6552352655987966212_!_110_!_news_military_!_四野在东北的武器装备是哪里来的？_!_
+6552277482186211853_!_110_!_news_military_!_退伍军人身上都有这3个“坏毛病”！百分之九十的人跑不掉_!_兵哥哥,军人,退伍军人,坏毛病,吹哨子
+6551967839236915726_!_110_!_news_military_!_无人机直接撞向美航母 这国凶狠手段连俄罗斯都自愧不如_!_伊朗,航母,F-15E,A-18E,空空导弹,RQ-170,战斗机,无人侦察机
+6552379904967573763_!_110_!_news_military_!_军事介入叙利亚超2年，军费超50亿，为什么普京却笑开了花？_!_
+6552320216016093448_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552439446715761156_!_110_!_news_military_!_美国有飞机墓场，而乌克兰有坦克墓场，但境遇大不相同_!_主力坦克,飞机墓场,坦克,美国,乌克兰
+6552469535096570382_!_110_!_news_military_!_对越反击战打草惊蛇战术妙，坑死越军1个加强排_!_掩蔽部,马旭旺,交通壕,手榴弹,越军
+6552273325781942787_!_110_!_news_military_!_国内海军驱逐舰发展沿革_!_人民海军,排水量,海军,防空导弹,相控阵雷达
+6552311625620128007_!_110_!_news_military_!_为什么亮剑里边赵刚大学刚毕业到部队就是团政委？_!_
+6552331203700064772_!_112_!_news_travel_!_探秘最后的香格里拉之魂，纵使错过全世界，也不愿意错过你_!_墨竹工卡,新都桥,川藏南线,加油站登记表,雀儿山,加油站实名登记小票,川藏线
+6552469598875156999_!_112_!_news_travel_!_平遥古城游记_!_古建筑古建筑,古建筑古建筑古建筑,平遥古城,古建筑,日昇昌
+6552344755538231815_!_112_!_news_travel_!_去北海道，有什么推荐新奇的温泉胜地吗？_!_十胜川,北海道,十胜川温泉,定山溪温泉,函馆市
+6552414899950060039_!_112_!_news_travel_!_周末去哪儿？来安徽感受国家级绝美风景道_!_白马尖,单龙寺,霍山石斛,霍山县,大别山
+6552448427626594829_!_112_!_news_travel_!_远行吧！来一场想走就走的旅行_!_
+6552389400028774926_!_112_!_news_travel_!_亿航在西安的无人飞机秀演砸了，但并不影响继续它的表演_!_西安市,吉尼斯世界纪录,红五月西安城墙国际文化节,李昌林,自然环境,无人机
+6552454141652763149_!_112_!_news_travel_!_登高投身云海间，早晨抚摸阳光炊烟，荡漾湖上与湖怪嬉戏_!_喀纳斯湖,湖怪,柯斯达,喀纳斯,观鱼台
+6552476438526689805_!_112_!_news_travel_!_长汀连城两日游，走过古城，探访培田古民居，登顶冠豸山_!_连城县,最美古村落,培田古民居,客家,冠豸山,长汀
+6552379385620464141_!_112_!_news_travel_!_尼山圣地孔子湖畔千人徒步 快乐行走欣赏尼山美丽风光_!_湿地公园,徒步,尼山镇,孔子,曲阜市
+6552164772580688131_!_112_!_news_travel_!_有哪些日漫中有浪漫的樱花场景？_!_
+6552485699130491395_!_112_!_news_travel_!_湖北宜昌：最浪漫的相亲节在薰衣草花海中举行！_!_普罗旺斯,花海,寻找意中人,帅哥美女,帅哥上台玩起互动游戏
+6552269903129215501_!_112_!_news_travel_!_泰山游记——一个人的自驾之旅_!_
+6552364941871940109_!_112_!_news_travel_!_到广西旅行这几个神秘秘境，让人身临其中，仿佛世界在脚下_!_天坑,百色市,通灵大峡谷,靖西县,漂流
+6552424427252875528_!_112_!_news_travel_!_为什么许多中国人喜好去泰国，而不去三亚？_!_
+6552349396862763527_!_112_!_news_travel_!_五一小长假首日：通道侗寨游人如织_!_五一小长假,文化旅游,游人如织,通道县,吊脚楼
+6551958358578954765_!_112_!_news_travel_!_靠花47亿维持的中国沙漠第一湖，旺季门票不下百元！_!_龙王宫,月牙泉,甘肃省,娘娘殿,敦煌市
+6552287191421157902_!_112_!_news_travel_!_旅行是一种爱好，了解新鲜的事物_!_三亚湾,冲浪,大东海,热带雨林,三亚
+6552295831205052936_!_112_!_news_travel_!_珠海国际会展中心斩获行业大奖 推动珠海市会展旅游产业加快发展_!_珠海国际会展中心,MICE China,奖旅游,沃尔玛,丝绸之路,国际博览会
+6552254353258840579_!_112_!_news_travel_!_贵州省遵义市湄潭县辖区内旅游景点简介_!_湄潭,康熙湄俯瞰湄潭全景潭县志,天下第一壶,平越,茶文化,湄潭县
+6552221393000333581_!_112_!_news_travel_!_从北京自驾去新疆怎么走合适？_!_
+6552130969510347022_!_113_!_news_world_!_作为俄罗斯总统，普京上台以来有哪些失误？_!_
+6552367905202241805_!_113_!_news_world_!_哈尔滨一弃婴被荷兰父母领养，16年后凭一张纸条回来寻亲，你怎么看？_!_
+6552472275822575885_!_113_!_news_world_!_怎么评价普京？_!_
+6552439468391923981_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552466444217483790_!_113_!_news_world_!_为什么说种姓制度制约了印度的发展？连阿米尔汗都承认这点_!_印度教,种姓制度,首陀罗,印度,贱民
+6552337945792086280_!_113_!_news_world_!_如果日本国土滑向海沟沉没，要求中国人民收留，作为中国人你会接受日本人吗？_!_
+6552369813862220291_!_113_!_news_world_!_美智库：中国科研经费飙升装备加速突破将损害美国优先_!_美国政府,美国优先原则,科研经费,战略与国际研究中心,美国
+6552480554443342093_!_113_!_news_world_!_希特勒和拿破仑都在俄罗斯折戟沉沙，难道俄罗斯真的不可战胜吗？_!_
+6552364382259511566_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552275817391456776_!_113_!_news_world_!_巴基斯坦两处煤矿发生矿难致18人死亡_!_巴基斯坦,矿难,奎达市,俾路支省,煤矿
+6552448014219215363_!_113_!_news_world_!_直击：美国历史上第一位平民黑人总统的成长经历_!_夏威夷,美国民主党,堪萨斯州,奥巴马,哥伦比亚大学
+6552476919265231118_!_113_!_news_world_!_侣行在伊拉克，叙利亚等的土豪行为，为后来的探险家造成巨大的伤害，您怎么看？_!_
+6552444415451857421_!_113_!_news_world_!_鲁哈尼：如果美国退出核协议，将会感到“前所未有”的后悔！_!_鲁哈尼,毫无意义,特朗普,伊朗,东方IC
+6552358932403192071_!_113_!_news_world_!_如果发达国家都在非洲，其他地区都是发展中国家的话，我们的审美是否会趋向认为黑种人比黄、白人种好看？_!_
+6552146982285083149_!_113_!_news_world_!_朝鲜5日零时起 启用新平壤时间 与韩国统一时区_!_朝鲜,时区,朝中社,金正恩,朝鲜半岛
+6552297287563870478_!_113_!_news_world_!_作为俄罗斯总统，普京上台以来有哪些失误？_!_
+6552455159631315213_!_113_!_news_world_!_为什么外国元首会见特朗普女儿伊万卡时都很开心？_!_
+6552340484428136967_!_113_!_news_world_!_和中国最友好的国家_!_哈萨克斯坦,吉尔吉斯斯坦,塔吉克斯坦,哈耶克,巴基斯坦
+6551896429760086532_!_113_!_news_world_!_读老报：99年肯尼迪二世坠亡事件，肯尼迪这个家族总出大事情_!_贝赛特,肯尼迪家族,马萨葡萄园岛,FBI,海岸警卫队
+6552388625957388552_!_113_!_news_world_!_为何奥巴马刚上台就获诺贝尔和平奖？他给世界带来和平了吗？_!_
+6551646006751002893_!_115_!_news_agriculture_!_玉米心叶展不开是什么原因导致的？_!_
+6552226193133797892_!_115_!_news_agriculture_!_老婆非要装马桶，农村我只习惯蹲坑！最后还是老妈有办法_!_Haier,洗澡刷,八件套,马桶盖,妇洗器
+6552361569248870919_!_115_!_news_agriculture_!_降低郁闭果园密度！烟台大樱桃老果园改造关键技术_!_烟台,樱桃,果园,樱桃树,烟台大樱桃
+6552154847863374350_!_115_!_news_agriculture_!_一个只有六户人家的偏远山村，只见到三户，看看都拍到了些什么？_!_运城市,女主人,秦四海,外孙女
+6552349984405061895_!_115_!_news_agriculture_!_怎样防治小麦秆锈病？_!_
+6552277351445561870_!_115_!_news_agriculture_!_90后小伙用区块链养鸡一年时间完成10万+会员认养吸引资本数千万_!_唐平冬,鸡笼,养鸡,区块链,摩拜单车
+6552422744871403789_!_115_!_news_agriculture_!_现在农村人为什么都不种地了？_!_
+6552413957380899335_!_115_!_news_agriculture_!_世界上最奇葩的水果，你未必都见过，吃过其中的一种我就服你_!_玉米,红香蕉,草莓玉米,水果,鸡蛋果
+6552274282032923149_!_115_!_news_agriculture_!_为什么早市蔬菜那么新鲜？_!_蔬菜,早市
+6552294471449444878_!_115_!_news_agriculture_!_农村葬礼，随了份子后菜没吃上，只喝了几口汤_!_几口,农村
+6552161256785052167_!_115_!_news_agriculture_!_还大热天在瓜棚照看瓜吗？有了这“黑科技”，随时随地查看地里瓜_!_午睡床,搓澡神器,硅藻土,单筒望远镜,望远镜
+6552298092408865283_!_115_!_news_agriculture_!_懒人适合种的果树：长得多、好打理，你吃过几种？_!_黄龙火龙果,杏子,李子,大红袍,无花果
+6552463379565380100_!_115_!_news_agriculture_!_成年牛要想放牧育好肥，掌握方法是关键_!_玉米面,放牧地,棉籽饼,青干草,放牧育肥
+6552375432849654285_!_115_!_news_agriculture_!_大胃王密子君直播吃重庆老火锅，服务员在旁边不停加菜，真心疼_!_服务员,大胃王,密子君,火锅,重庆火锅
+6552290207910068739_!_115_!_news_agriculture_!_产蛋率有下降了，高手有几点指导意见_!_大肠杆菌病,产蛋率,消毒剂,鸡舍
+6552475323152204292_!_115_!_news_agriculture_!_农村小伙教你几招：轻松挑选出优质的东北大米_!_东北大米,支链淀粉,蛋白质,米粒,大米
+6552306181845549315_!_115_!_news_agriculture_!_水泥钢筋价格大幅度的上涨，农村自建房怎么办？_!_
+6552339068653404676_!_115_!_news_agriculture_!_现在农村都兴种这种果树了，好打理、易结果，一年都不吃完_!_谷酰胺,山楂树,郁金香,土库曼斯坦,蓝莓
+6552431726319108356_!_115_!_news_agriculture_!_为什么果树旺长不结果？为什么果树结果很少？如何改造成高产树？_!_
+6514205024938098958_!_116_!_news_game_!_《王者荣耀》韩信偷鲲这个梗是从哪里来的？_!_韩信,王者荣耀,韩信偷鲲,庄周,荣耀分析师
+6551897008574038535_!_116_!_news_game_!_《DNF》玩家感叹；当时迫不及待头脑发热了一下，现在后悔莫及！_!_dnf
+6552285136660988174_!_116_!_news_game_!_LOL每天后半夜都会有自动送人头的，都是0-几的，查看战绩都是连输的，这是为什么？_!_
+6552473493194146312_!_116_!_news_game_!_UZI直播黄金分段，却被国服路人单杀！玩上单薇恩遭疯狂制裁！_!_UZI,薇恩,Ming,德莱文,诺手
+6552082887972225540_!_116_!_news_game_!_我们快被群里的小程序游戏逼疯了！分享几个解决办法……_!_微信游戏,小游戏,忍者神龟,游戏
+6552468310510797319_!_116_!_news_game_!_LPL冠军UZI虎牙直播首秀 弹幕人气排行双双第一！400万爆炸人气！_!_LPL,直播间,UZI,MSI,uzi
+6552393563148124685_!_116_!_news_game_!_大吉大利，晚上吃鸡，来看你是那种吃鸡选手！_!_类玩家,游戏,快递员
+6552378234917356039_!_116_!_news_game_!_王者荣耀：玩家炫耀15万金币账号，却被网友一张图打脸扎心了！_!_金币,机关术,游戏,豹子头,玩家
+6552459783343964424_!_116_!_news_game_!_《魔兽争霸3》的职业选手，能打赢两个业余选手吗？_!_
+6552460315865383175_!_116_!_news_game_!_DNF：在游戏中最让人无奈的事情有哪些？_!_
+6552436941210518023_!_116_!_news_game_!_梦幻西游：掉须弥只是逆袭的开始？喜闻乐见的结局_!_梦幻西游,5技能,9技能
+6552261974535700750_!_116_!_news_game_!_《空洞骑士》值得入手吗？_!_
+6552246460052668936_!_116_!_news_game_!_DNF：魂牛记录再次刷新，剑魂携驱魔弹药奶爸47层通关成功！_!_驱魔,47层,剑魂,魔兽副本,超大陆
+6552378058580427011_!_116_!_news_game_!_吃鸡游戏里一百个人里哪些是人机？_!_
+6552392702023958797_!_116_!_news_game_!_爆爆，呆呆，阿木木历，汤米，清水，暗墨，星星知多少，大海，火焰，楚河…你们更爱谁？_!_
+6552485879582032398_!_116_!_news_game_!_王者荣耀：5个“小学生”最爱的英雄，进去就秒选的要小心了_!_亚瑟,狄仁杰,小学生,后羿,韩信
+6552360074893853197_!_116_!_news_game_!_互粉有效性和原创收益实测分析，大家根据实际情况进行选择！_!_头条号,粉丝
+6552382400649757191_!_116_!_news_game_!_王者荣耀：妹子刚上铂金就被封号，理由很搞笑，网友却直呼好傻_!_刷金币,铂金段位,铂金,封号,妹子
+6552478452266566158_!_116_!_news_game_!_MSI入围赛第一支晋级战队诞生！中国队的老朋友_!_俄罗斯,GPL,GMB,战队,MSI
+6552386978669658632_!_116_!_news_game_!_《碧蓝航线》舰娘120级排行榜 舰b120级PVE节奏榜_!_姐妹船,舰娘,PVE,120级,需求能力
+6552201057550729475_!_101_!_news_culture_!_如果能带五本书穿越到古代，你会带哪五本？为什么？_!_
+6552418032864461059_!_101_!_news_culture_!_如何在社会中成长？_!_
+6552389747883376909_!_101_!_news_culture_!_下联：饱暖思淫欲。请赐上联？_!_
+6552450531242017032_!_101_!_news_culture_!_迷失在历史长河中的陕西茯茶文化还能归位吗？_!_
+6552370157333774852_!_101_!_news_culture_!_福建茶企举办全球品鉴会 首站走进厦门_!_筼筜书院,茶文化,华祥苑国宾茶,中国茶叙外交,茶叙外交,王旭烽
+6552345593061048589_!_101_!_news_culture_!_特朗普跟里根有可比性吗？_!_
+6552400131654680845_!_101_!_news_culture_!_文化产业管理专业就业前景如何？_!_
+6547925235021119758_!_101_!_news_culture_!_金庸笔下的男主角谁最厉害？_!_
+6552416908342198542_!_101_!_news_culture_!_上联：星繁风幽催叶梦，下联你怎么对？_!_
+6552128603117584653_!_101_!_news_culture_!_上联：池畔蛙鸣，应知红短绿长，怎么接下联？_!_
+6552228068151263502_!_101_!_news_culture_!_上联是：柳映池塘鱼上树，如何对下联？_!_
+6552356222215913988_!_101_!_news_culture_!_论雅俗共赏，我只服他_!_柳牛图,中国名画家全集-齐白石,白石老人自述,墨虾,齐白石作品珍藏卷,芙蓉,鹤寿图,仙鹤,齐白石,贝叶,大写意,蛙声十里出山泉,白石老人,贝叶蜻蜓,齐白石全集·第一卷,白石,桃实图
+6552266708755677700_!_101_!_news_culture_!_秦腔流派介绍：衰派，马连良称赞他是“衰派老生中的一绝”_!_衰派老生,易俗社,秦腔,马连良,临潼县
+6552482679923671310_!_101_!_news_culture_!_河南话和广东话哪个更像是古汉语？_!_
+6552168179630604814_!_101_!_news_culture_!_黎伟明将与法意艺术家在深圳联展_!_深圳雅昌艺术馆,马克·兰波,·兰波,马克·,艺术家
+6552315996021981710_!_101_!_news_culture_!_晋唐小楷八种《唐刻宋拓晋唐小楷八种》大图欣赏，清代集本_!_小楷,褚遂良,王羲之,东晋,乐毅论
+6552364863409094925_!_101_!_news_culture_!_在和女生聊天的时候，如何快速找到话题？_!_
+6552384975134523918_!_101_!_news_culture_!_正能量 正楷标准教程4_!_
+6552408920311202317_!_101_!_news_culture_!_关于《红楼梦》中的那些经典词句，你读懂了多少？_!_红楼梦
+6552416586274177544_!_102_!_news_entertainment_!_刷爆朋友圈的综艺回归！《朗读者Ⅱ》有了新变化……_!_朗读者,贾平凹,宗庆后,清华大学,薛其坤,中国篮球协会,朗读者Ⅱ
+6552450352333980168_!_102_!_news_entertainment_!_《昆池岩》开启恐怖片热潮，今年的十大海外惊悚片你最期待哪部？_!_鸡皮疙瘩,昆池岩,人皮客栈,恐怖片,孤堡惊情,招魂2,分裂,阴风阵阵,恐怖电影,温子仁,七年之夜,女巫,招魂,拍栗得,寂静之地
+6552341111682105863_!_102_!_news_entertainment_!_快乐大本营谢娜回归众人都发文欢迎 为什么独独吴昕没有动静？_!_吴昕,李维嘉,谢娜,杨迪,快乐大本营,快乐家族
+6552319781154849288_!_102_!_news_entertainment_!_盘点漫威电影人气最高10位女主角：黑寡妇排第10 第一名是她_!_10第一,女主角,黑寡妇排,惊奇队长,漫威电影人气
+6552477112601674253_!_102_!_news_entertainment_!_那些怀孕还坚持拍戏的女明星，蔡少芬最敬业，她就有点尴尬了！_!_怀孕,胡杏儿,孤芳不自赏,怀有身孕,公众人物,那年花开月正圆,小时代：刺金时代,蔡少芬
+6552379844989026829_!_102_!_news_entertainment_!_林嘉绮出席发布会，网友：她的每一次转身，就没让我们失望过_!_林嘉绮,林嘉绮出席发布会
+6552264730453475847_!_102_!_news_entertainment_!_5步计划开始你的演艺事业_!_表演课程,迈斯纳,电影,独立电影,演艺事业
+6552425051000406542_!_102_!_news_entertainment_!_何老师希望谢娜下辈子是他的“路人”_!_谢娜,何炅,下辈子,土鸡蛋,生日快乐
+6552294600445264388_!_102_!_news_entertainment_!_「王源壁纸」就喜欢看你那纯真的笑容_!_
+6552467409620435464_!_102_!_news_entertainment_!_当红女星的十八岁：迪丽热巴古力娜扎参加虞姬海选，范冰冰当丫鬟_!_古剑奇谭,还珠格格,南溪斋笔录,丁建国,克拉恋人,范冰冰,王的盛宴,古力娜扎,宝莲灯,缝纫机乐队,金粉世家,刘亦菲,赵灵儿,喜剧之王
+6552388102906708237_!_102_!_news_entertainment_!_公子扶苏为何心甘情愿自尽？_!_
+6552421804667830788_!_102_!_news_entertainment_!_杨幂黑色礼服秀性感香肩，精修图温婉动人，无精修的你觉得怎样？_!_性感,精修图,礼服,杨幂,北京大学生电影节
+6552466802226496008_!_102_!_news_entertainment_!_用“女儿裸照”赚钱，这种也配称之为“妈妈”？_!_安娜,薇莉塔,汉娜,花花公子,艳娃传,春之森林,小公主,于佩尔
+6552469496739660292_!_102_!_news_entertainment_!_孔令辉说出当年分手实情，马苏成“万人睡”, 数十名男星被牵连_!_马苏,孔令辉,李小璐,黄毅清,魏大勋
+6552341998349582856_!_102_!_news_entertainment_!_谢娜创吉尼斯纪录，你知道是哪两项吗？_!_吉尼斯世界纪录,粉丝,吴昕,快本,快乐大本营,谢娜,快乐家族
+6552256431398060547_!_102_!_news_entertainment_!_他曾红过释小龙，误入歧途坐牢10年，如今洗心革面重新做人！_!_释小龙,颜正国,误入歧途,天才童星,洗心革面
+6552276831771296263_!_102_!_news_entertainment_!_《彼得兔2》定档 萌贱小兔2020年2月7日将归来_!_碧翠丝,票房,彼得兔2,摄影棚,比得兔,索尼公司,奇幻森林
+6552396401978901006_!_102_!_news_entertainment_!_49岁邓文迪，网友：霸气侧漏，如此汹涌澎湃！_!_汹涌澎湃,霸气,霸气侧漏,邓文迪
+6552133237236302088_!_102_!_news_entertainment_!_一个人的微博账号透露了哪些信息？_!_
+6552431568378397197_!_103_!_news_sports_!_四国青年女篮亚特拉斯赛在秀举行_!_匈牙利,篮球,对抗赛,匈牙利队,拉脱维亚
+6552251683378823684_!_103_!_news_sports_!_史蒂文斯神奇调整凯尔特人加时再胜76人，苏群：76人完了_!_贝里内利,凯尔特人,76人,史蒂文斯,霍福德
+6552246563991716100_!_103_!_news_sports_!_大连海那么多，都有什么海上游乐项目？_!_
+6552317618647204356_!_103_!_news_sports_!_德罗赞连续13场季后赛面对骑士没有投进三分_!_德罗赞,季后赛,德玛尔
+6552329149871030542_!_103_!_news_sports_!_亚洲杯分组国足抽得上上签，国足就一定能出线吗？_!_
+6552280342454075917_!_103_!_news_sports_!_撒彩纸也不看清楚，打扫了足足七分钟！凯尔特人窃喜中！_!_凯尔特人,恩比德,76人队,塔图姆,霍福德
+6552281514569105672_!_103_!_news_sports_!_时间过去将近两年了，大家是如何看待杜兰特加盟勇士这件事情？_!_
+6552258575102640647_!_103_!_news_sports_!_卡拉斯科的底线过人和本泽马的哪个更优秀？_!_马德里竞技,中超联赛,广州恒大,本泽马,卡拉斯科
+6552413223063126280_!_103_!_news_sports_!_NBA如果用带队能力给现役球星排名，前五都有谁？_!_
+6552132778392027655_!_103_!_news_sports_!_延边富德怎么了？主帅朴泰夏怎么了？_!_延边富德,梅西,朴泰夏,女足,长春
+6552839162100187662_!_103_!_news_sports_!_米体：比格利亚将在意大利杯决赛中复出_!_阿根廷,维罗纳,意大利杯决赛,比格利亚,贝内文托
+6552474943781601806_!_103_!_news_sports_!_男乒世乒赛九连冠_!_世乒赛,九连冠,德国,刘国梁,男乒
+6552143418888290564_!_103_!_news_sports_!_郭跃在社交平台上说丁宁，每次输掉比赛都说受伤，对此你怎么看？_!_
+6552317499189231875_!_103_!_news_sports_!_猛龙主教练跟泰伦卢在这个系列赛的执教水平你怎么看？_!_
+6552359913039856132_!_103_!_news_sports_!_5战115分，隔扣被犯规还“口出狂言”，球迷：听这话就有巨星潜质_!_塔图姆,垃圾话,米切尔,费城76人,凯尔特人
+6552351022725988611_!_103_!_news_sports_!_你们觉得李秋平加盟山东队的可能性大吗？_!_
+6552839148892324359_!_103_!_news_sports_!_亚洲杯小组赛中国队赛程出炉：首战吉尔吉斯斯坦_!_扎耶德体育场,菲律宾,阿联酋,阿布扎比,吉尔吉斯
+6552350810351599880_!_103_!_news_sports_!_cba历史上最差的小外援是谁？_!_
+6552274322960941575_!_103_!_news_sports_!_姚明的“奥运第四问”：盼00后运动员继续为国旗奋斗_!_中国篮协,运动员,奥运三问,朗读者,姚明,篮球运动员
+6552282971049558286_!_103_!_news_sports_!_女排欧冠半决赛，瓦基弗银行3:2险胜科内利亚诺，你怎么看？_!_
+6552281264852828685_!_104_!_news_finance_!_黄金原油下周走势分析_!_原油
+6552343939049849101_!_104_!_news_finance_!_一斤两块五的青菜人都买的吃，为什么两块五的股票没人买呢？_!_
+6552469199271232014_!_104_!_news_finance_!_上证指数目前已形成2次的超大单资金的底背离了，已有条件见底了_!_成交额,大唐电信,巴菲特,孙正义,观察线,上证指数
+6552430268475506948_!_104_!_news_finance_!_为什么《英雄联盟》国服皮肤卖那么贵？_!_
+6552326526963024391_!_104_!_news_finance_!_交通银行东方航空信用卡申请难吗？白金卡比较难入手_!_航空出行需求,白金卡,申请,银行东方航空信用卡,交通银行,东方航空
+6552467954926092807_!_104_!_news_finance_!_环保中小企业享受减税利好，环保装备制造企业同步起飞_!_环保装备制造业,我国环保装备制造业,增值税,营改增,规模纳税人
+6552431024020652548_!_104_!_news_finance_!_力帆股份再启定增 拟募资24亿元投新能源汽车产业链_!_电动车,锂电芯,力帆股份,能源站,智能新能源汽车
+6552442691706159373_!_104_!_news_finance_!_震荡行情和单边行情如何区分？_!_
+6552251579871789581_!_104_!_news_finance_!_宿城举办现代服务业招商推介会 28个项目成功签约_!_推介会,宿城,三产兴城,服务业,中央商场
+6552468628803944973_!_104_!_news_finance_!_龚方雄：BAT回归将加速中国资本市场的机构化和国际化_!_龚方雄,抵押品,摩根大通,北京大学,中国资本市场
+6552339048877261064_!_104_!_news_finance_!_普通外汇投资者该如何炒外汇？_!_
+6552253890295759107_!_104_!_news_finance_!_宋鸿兵的《货币战争》这本书可信度有多高？_!_
+6552423809478033924_!_104_!_news_finance_!_两市晚间公告一览：一公司24.8亿投新能源汽车！股民：这够吗_!_电动车,新能源汽车
+6552366946933801485_!_104_!_news_finance_!_黄金周评：非农意外利多黄金维持震荡，多头发力上方阻力重重_!_美国,失业率,多头,重点关注,非农
+6552455207815479822_!_104_!_news_finance_!_深度好文：中国制造业需要摆脱三个陷阱_!_附加值,李哲贤,经济模式,制造业
+6552491711287787780_!_104_!_news_finance_!_关于均线系统，有什么独特的、有效的见解跟设置用法吗？_!_
+6552364125538746893_!_106_!_news_house_!_西安房价还会涨吗？几方面分析西安房市水平！_!_房价,武汉,大数据,西安市,棚改货币化
+6552273262766719245_!_106_!_news_house_!_2018年下半年的房价还会涨吗，何时入手是最佳时期？_!_
+6552031238398411267_!_106_!_news_house_!_贷款买房要趁早！下月利率有望迎来“16连涨”_!_房价,高房价,买房,房贷利率,利率上涨
+6552135814120210695_!_106_!_news_house_!_住房公积金贷款如何申请到高额度？_!_
+6552334474917445896_!_106_!_news_house_!_石家庄买房可以买样板房吗？样板房到底能不能买？_!_
+6552460291588751624_!_106_!_news_house_!_五月份是买房的好时间吗？_!_
+6552302509057114372_!_106_!_news_house_!_昌黎县房价怎么样？_!_
+6552043051609489934_!_106_!_news_house_!_一起看房的哥们说房价1万太扯淡了，后来四万多一平，哥们蒙圈了_!_南三环,房价,北京,小房子,买房
+6552433957302960397_!_106_!_news_house_!_日本楼市泡沫到后期是怎样一副光景？_!_
+6552346823514652935_!_106_!_news_house_!_郑州房产中介到底有多坑爹？_!_
+6552290644356760068_!_106_!_news_house_!_拆迁中，理由一定要实事求是，以免弄巧成拙_!_小产权,镇政府,拆迁,宅基地,商品房
+6552403915034329607_!_106_!_news_house_!_南京六合区地铁口别墅！金牛湖嘉恒有山总价180万起！嘉恒有山_!_度假区,400-060-1618转33361,南京,六合区,金牛湖
+6552458405238276611_!_106_!_news_house_!_这些城市，正在拿命欢迎你_!_双轨制,落户,高级工,职业资格,成都市
+6552292298632200717_!_106_!_news_house_!_瞭望刊文：房地产税改革渐行渐近，还需突破四个阻力区_!_史耀斌,税改革,土地财政,瞭望,不动产,房地产税
+6552375430064636429_!_106_!_news_house_!_ANZ用五张图大胆预测：澳大利亚房价在今年会停止下跌并开始上涨_!_墨尔本,中央银行,房产,房产市场,澳大利亚,房价
+6552415010939732227_!_106_!_news_house_!_未来房价会跌到什么程度？_!_
+6552473682587943182_!_106_!_news_house_!_买房银行审批通过但就是不放款，怎么办？_!_
+6552263560813085197_!_106_!_news_house_!_房价终于跌了？“深圳房价19连跌”_!_黄鹤楼,深圳,深圳新房,商品住宅,资源委员会
+6552460461692944903_!_107_!_news_car_!_2018年，东风商用车这些事值得关注！_!_东风天龙,L4,东风商用车,东风,HOWO,无人驾驶,V2V
+6552324635537441038_!_107_!_news_car_!_如果给你三十万，你会选择什么车？_!_
+6552430127643361799_!_107_!_news_car_!_雷克萨斯科幻车亮相，造型更激进，剑指宝马X6_!_宝马X6,日内瓦车展,GLECoupe,雷克萨斯,SUV
+6552460858440548872_!_107_!_news_car_!_别被大屏、天窗所迷惑，不管高配、低配，这些车载配置不能省_!_阴雨天,日间行车灯,转向过度,车身稳定控制系统,车载
+6552458420534903048_!_107_!_news_car_!_预算十万元买什么SUV好？_!_
+6552345045792456973_!_107_!_news_car_!_国产沃尔沃，安全性能第一，故障率几乎为0，为何没人买？_!_
+6552455573227438605_!_107_!_news_car_!_延续王者风范，全新一代奔驰G63正式亮相_!_新一代,G63,王者风范,澳大利亚
+6552380007728021764_!_107_!_news_car_!_雅阁，凯美瑞，天籁后期费用，大约都多少钱？_!_
+6552303649492566542_!_107_!_news_car_!_Honda为中国市场打造专属车型亮相上海，众多焦点车型展出！_!_X-ADV,CBR1000RR,NS110i,Africa,新大洲本田,Honda,Twin,CB190R,CBF190R
+6552401233238295054_!_107_!_news_car_!_新能源汽车展在南京9月14日举办汽车行业盛宴_!_新能源,能源汽车展,汽车行业,新能源汽车,新能源汽车展
+6552440847915286788_!_107_!_news_car_!_只有6万现金，想买个合资车型，有没有可以推荐的？_!_
+6552444348703703559_!_107_!_news_car_!_从遭人嫌弃到最具黑马潜质！这款SUV终于要翻身了_!_设置装备安排,皮尺部,触控式,众泰,T700
+6552486949976801806_!_107_!_news_car_!_5秒破百/纯电续航44km，新款保时捷卡宴插混版曝光_!_Cross,Turismo,Mission E,保时捷,Cayenne,新能源汽车,E-Hybrid
+6552264555664245252_!_107_!_news_car_!_十年车龄的蓝色日产370Z，夜惑街头！_!_VVEL,日产370Z
+6552440018986598926_!_107_!_news_car_!_开阳拉力赛赛道封路不要怕，赛道消息抢先知~_!_拉力赛,赛道
+6552349681836360199_!_107_!_news_car_!_为你的情怀买单！JEEP全新指南者震撼来袭，霸气侧漏，售价感人_!_CR-V,Jeep指南者,JEEP,SUV,Jeep,全新指南者,指南者
+6552303053679100430_!_107_!_news_car_!_虽是冷门但也是好车，居然也有大众！_!_定速巡航,SYNC,福特,帕萨特,发动机
+6552324377428361731_!_107_!_news_car_!_夏天来了，钥匙别往裤兜塞，试试这6款多功能钥匙包，实用有格调_!_简单零钱包卡包,钥匙包,多功能
+6552272380138357256_!_107_!_news_car_!_5月6日09:23，津沧高速驶往沧州方向K47处的事故处理完毕；京哈高速驶往哈尔滨方向K81处的事故仍在处理中_!_高速公路,天津,哈尔滨
+6552361156177035780_!_107_!_news_car_!_汽车暴晒像蒸笼？越来越多司机装“太阳伞”，车内比开空调还凉爽_!_司机,太阳伞
+6552420339438060039_!_108_!_news_edu_!_今日就教大家复原魔方的方法_!_
+6552380233704538638_!_108_!_news_edu_!_地理考试加油！_!_
+6552402863065137667_!_108_!_news_edu_!_婺州家庭教育大讲坛暨环小暖爸活动在中国婺剧院举行_!_华东师范大学,环城暖爸,文明之歌,家庭教育,环城小学,一起来金华,婺剧院,醉炫琴韵
+6552453922559099400_!_108_!_news_edu_!_这样的老师最受学生欢迎！_!_
+6552311349475541517_!_108_!_news_edu_!_教师通宵批改试卷猝死家中，其妻申请工伤或法院支持告诉我们什么_!_非正常死亡,脑力劳动,编制数,应试教育,教师
+6552467344646472205_!_108_!_news_edu_!_幽默大学第2课：幽默基本公式之连接词_!_连接词
+6552346158956544516_!_108_!_news_edu_!_北大最有名的建筑之一，共产党早期组织在此成立，影响毛主席一生_!_北京大学,蔡元培,钱玄同,北大,京师大学堂,新青年
+6552378848917324292_!_108_!_news_edu_!_李永培：课程方案调整、高中教材变革将倒逼高考改革进一步深化_!_三位一体,课程设计,李永培,高考改革,浙江省
+6552450606110343683_!_108_!_news_edu_!_沪上中小学今年第一波毕业照出炉啦，你也晒晒自己的吧！_!_上海市,毕业照
+6552340481781531149_!_108_!_news_edu_!_石家庄铁路职业技工学校2018年招生简章_!_客运段,邯黄铁路,技工学校,上海地铁,天津地铁
+6552130501967086083_!_108_!_news_edu_!_女子科目二满分网上却显示零分 工作人员：只能再考_!_监控平台,监管平台,管理所,科目,长春市
+6552331086767063304_!_108_!_news_edu_!_现在的职业学校能真正学到技术吗？_!_
+6552410076773089799_!_108_!_news_edu_!_短文 回忆高三_!_高三
+6552329452058050820_!_108_!_news_edu_!_高数难学吗？_!_
+6552131120245244168_!_108_!_news_edu_!_OPPO成功的秘诀是什么？有什么可以学习的吗？_!_
+6552481254074220808_!_108_!_news_edu_!_美国是怎样进行爱国主义教育的？_!_
+6552386823505576195_!_108_!_news_edu_!_高考选专业是自己决定还是由父母决定好？_!_
+6552378958287995140_!_108_!_news_edu_!_快中考了，父母怎样配合孩子能考出个好成绩？_!_
+6552400550288163079_!_109_!_news_tech_!_能进阿里巴巴集团做管理层领导的人，出来创业成功率能达百分之八十吗？_!_
+6552284320730448132_!_109_!_news_tech_!_除了华为、阿里巴巴以外，你还知道哪些称霸世界的中国高新企业？_!_
+6552391696343106056_!_109_!_news_tech_!_腾讯没有梦想_!_刘炽平,腾讯,恋与制作人,腾讯传,SNG,光荣使命,马化腾,任宇昕,旅行青蛙,腾讯微博,微视,短视频,腾讯视频,张小龙
+6552359702649373191_!_109_!_news_tech_!_OTT只能卖开机？酷开网络用活动让广告“玩”起来_!_广告创意,酷开网络,酷开系统,广告形式,创维
+6552355069688283400_!_109_!_news_tech_!_高度发达的外星人有复活死人的科技吗？_!_
+6552366816449004040_!_109_!_news_tech_!_小程序逐渐成为人们生活一部分，微信生态规则与流量玩法 资源_!_微信,马云,流量,支付宝
+6552307873945223683_!_109_!_news_tech_!_《2018年春季深圳人才供需与流动报告》发布，平均42人竞争一个职位！_!_深圳,深圳人才,春季深圳,报告,区块链,求职者
+6552345078726131975_!_109_!_news_tech_!_店铺开了快两个月了，换类目后每天一两个访客不知道怎么办？_!_
+6552273749595390478_!_109_!_news_tech_!_2018.5.6币值排名前20 总市值突破30000亿_!_总市值
+6552229877133607431_!_109_!_news_tech_!_如何做到更好的设计、更高的质量、更低的价格、更多的收益？_!_互联网,价值链
+6552474143894274308_!_109_!_news_tech_!_新买的电脑只有一个C盘，如何增加分区？_!_
+6552411484943548941_!_109_!_news_tech_!_芝麻信用分又更新了 马云：给600分以上福利！你芝麻分够吗？_!_芝麻分,支付宝,芝麻信用,信用分,马云
+6552425459265569037_!_109_!_news_tech_!_是充电一夜对手机损害大还是电量没用完就充电损害大？_!_
+6552429172440957444_!_109_!_news_tech_!_支付宝-好医保VS微信-微医保_!_微信,微医保,保险市场,奥特曼,支付宝
+6552333422973419784_!_109_!_news_tech_!_谁有好的微信头像可以分享？_!_
+6552394606397358339_!_109_!_news_tech_!_有哪些我们很少关注的国家，却有着全球前列的科技产品？_!_
+6552415641087770888_!_109_!_news_tech_!_在淘宝上面做分销商，怎么发货？_!_
+6552338041384468749_!_109_!_news_tech_!_“国民技术”的产品都应用在什么地方？_!_
+6552128599271407876_!_109_!_news_tech_!_微信小程序开发需要用到哪些？_!_
+6552280473840648456_!_110_!_news_military_!_有人说日本要是不参与第二次世界大战，现在世界上最强大的国家非日本莫属，你怎么看？_!_
+6552453638093013256_!_110_!_news_military_!_第一次车臣战争，俄罗斯面对小小的车臣为什么损失惨重？_!_
+6552278203098988808_!_110_!_news_military_!_二战时候的日本轰炸机一次带多少炸弹？_!_
+6552285030389907719_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552289699279405325_!_110_!_news_military_!_我国古代就有火炮，为什么没有继续微型化发展出火枪？_!_
+6552309217670201608_!_110_!_news_military_!_不考虑核武器，只是常规武器，美国多久能打赢俄罗斯？_!_
+6552361528698339848_!_110_!_news_military_!_央视聚焦国产航母诞生地，位于武汉的这个研究所实力刷屏_!_直升机,潜艇,导弹驱逐舰,国产航母,辽宁舰
+6552285197444841742_!_110_!_news_military_!_全程2000公里，历时5年的京九铁路为何才耗资507亿？_!_
+6552426298721960196_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552284219148599566_!_110_!_news_military_!_为什么这么多年阿富汗总是战乱不断？_!_
+6552434372245455364_!_110_!_news_military_!_特朗普从叙撤军？原来是条阴险的诡计！_!_特朗普,叙利亚政府军,逊尼派,大马士革,什叶派
+6552342634201874695_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552260625836278276_!_110_!_news_military_!_“中国标准”名扬苏丹维和任务区_!_中国首支维和,达尔富尔,首支维和直升机分队
+6552332698080248068_!_110_!_news_military_!_联合国安理会改革废除一票否决权，会导致什么后果？_!_
+6552438924889817358_!_110_!_news_military_!_为什么从1969年登月成功后再也没尝试登月? 在怕什么？_!_
+6552063551509364999_!_110_!_news_military_!_印度建造的第一艘国产航母，为何先后用了20多年时间？_!_
+6552284650264330759_!_110_!_news_military_!_19岁的对越反击战烈士：如果有机会上战场，我一定冲在最前面_!_王建川,老山战役,麻栗坡,砚山县,决心书
+6552489020515942660_!_110_!_news_military_!_人被原子弹炸死时，到底有多痛？_!_
+6552452487767392525_!_110_!_news_military_!_如果以色列与伊朗发生战争，谁赢的可能性最大？_!_
+6552272985430950407_!_112_!_news_travel_!_“东方莱茵河”竟然藏在京郊这座又大又美的公园内！关键免费玩！_!_森林公园,北京市,篱笆房,延庆,东关站
+6552363216788259086_!_112_!_news_travel_!_一辈子就一次的大四毕业，该选择旅行还是抓紧时间实习？_!_
+6552364919411441927_!_112_!_news_travel_!_为什么许多中国人喜好去泰国，而不去三亚？_!_
+6552143078449217806_!_112_!_news_travel_!_来文山玩，可以带什么特产？_!_
+6552455164530262542_!_112_!_news_travel_!_全国十大重点产茶县之一“寿宁”_!_寿宁县,景色寿宁县城景色寿宁县,产茶县,高山茶,景色
+6552244440444961283_!_112_!_news_travel_!_3个月贡献392亿，中国游客去日本却不为旅游观光，到底为的什么？_!_奢侈品,日本,忍野八海,旅游观光,导游
+6552359795070861831_!_112_!_news_travel_!_全国这8个5A级别的旅游古城镇，你去过几个？_!_周庄,国家5A级旅游景区,古镇,同里,文物保护单位,门市价,台儿庄
+6552331837639753987_!_112_!_news_travel_!_南宁城市怎么样？适合东北人生活吗？_!_
+6552471288714101255_!_112_!_news_travel_!_这就是西藏！_!_日喀则,西藏,然乌湖,绒布寺,桑耶寺
+6552424837086708238_!_112_!_news_travel_!_除一些必备物品外，旅行时还有哪些省空间少重量又实用的东西？_!_
+6552321020722676228_!_112_!_news_travel_!_贵州六盘水的这几个景点，错过一个你都会觉得可惜！_!_珙桐,六盘水,喀斯特,六盘水市,乌蒙山
+6552154233175540231_!_112_!_news_travel_!_他们将家猪丢弃在无人岛，几年之后，它们的生活竟是这样的！_!_巴哈马群岛,巴哈马,繁衍生息,日光浴
+6552393711643263496_!_112_!_news_travel_!_来四川旅游必去的八大景点_!_大熊猫,杜甫,峨眉山,色达县,峨眉郡志,乐山大佛
+6552389101813760259_!_112_!_news_travel_!_千姿百色弄秋光，求对子？_!_
+6552245876306215438_!_112_!_news_travel_!_为什么要来这里看映山红？北京这处抗战圣地有着悲壮的经历_!_
+6552267893738832398_!_112_!_news_travel_!_来一场说走就走的旅行，陕西宝鸡红河谷走起_!_河底村,营头镇,太白山,红河谷,眉县
+6552463286225338632_!_112_!_news_travel_!_四川游客张某带着六名女子在三亚沙滩烧烤被撒沙，这事你怎么看？_!_
+6552304224569393677_!_112_!_news_travel_!_农村集市上这些的东西，真的是便宜又好吃，简直赚大了_!_集市,农村集市,农村
+6552178256966058504_!_112_!_news_travel_!_全世界最出名的猫，它数第二，没人敢数第一了吧_!_加菲猫,好莱坞,狗品种,异国短毛猫
+6551491189198029316_!_112_!_news_travel_!_易县最美的五月，蓝色油菜花开好了_!_易县,狼牙山,花海,油菜花
+6552226122975674638_!_113_!_news_world_!_假如日本沉没，日本冲绳人迀移中国你会接爱吗？_!_
+6552459436139479555_!_113_!_news_world_!_108岁老人还下地干活，捉100只虾给家人做菜吃，还会喂给儿孙吃！_!_虾仁,印度
+6552318498788344078_!_113_!_news_world_!_美国建国不过两百年，为什么会成为世界强国？_!_
+6552270365001777668_!_113_!_news_world_!_这个13岁女孩从荷兰来到萍乡，只为全城寻找她的亲生父母！_!_萍乡,13岁女孩,晶晶,亲生父母,荷兰
+6552423223051419912_!_113_!_news_world_!_如果美国各个州独立了，世界格局会发生什么变化？_!_
+6552454947382755848_!_113_!_news_world_!_全球7对另类“神奇情侣”大比拼，让人啧啧称奇！_!_Katie,侏儒症,Charity,Arin,吸血鬼
+6552485388844269837_!_113_!_news_world_!_2018年美国中期选举，你认为特朗普会下台吗？_!_
+6552422469519540488_!_113_!_news_world_!_厨师的出路在哪里？_!_
+6552428749642531335_!_113_!_news_world_!_印度 女孩遭15人绑架/村委会：罚两人做俯卧撑_!_强奸案,性侵犯,强奸,印度,莫迪
+6552316767996215560_!_113_!_news_world_!_美国究竟发达到什么程度了？_!_
+6552445612715933959_!_113_!_news_world_!_如果美国各个州独立了，世界格局会发生什么变化？_!_
+6552419386114703619_!_113_!_news_world_!_美国建国不过两百年，为什么会成为世界强国？_!_
+6552491187465355523_!_113_!_news_world_!_纠结七天的日本行程，关西进出该不该去东京？_!_
+6552218782532632840_!_113_!_news_world_!_印度能够复制中国奇迹，超过日本成为世界第三大经济体吗？_!_
+6552325218914795783_!_113_!_news_world_!_如果日本国土滑向海沟沉没，要求中国人民收留，作为中国人你会接受日本人吗？_!_
+6552162349556433422_!_113_!_news_world_!_和普京同登《时代》周刊 杜特尔特：我可不是强人_!_时代,杜特尔特,菲律宾,欧尔班,普京,菲律宾总统
+6552437564798664967_!_113_!_news_world_!_哈马斯不承认以色列，那么哈马斯还能坚持多久？_!_
+6552070779897381389_!_113_!_news_world_!_美国人问：非洲人如何看待中国在非洲的投资？某网民：我们很支持_!_中国投资,铝土矿,铂族金属,非洲人,只想告诉你
+6552284574586503428_!_113_!_news_world_!_能够连接印度洋和太平洋的泰国克拉运河项目是否已经开工，一旦建成对新加坡有影响吗？_!_
+6552407759990555144_!_115_!_news_agriculture_!_中国发展得有多快，你知道吗_!_互联网,衣锦还乡
+6552362877179658504_!_115_!_news_agriculture_!_重庆黄葛树种植基地在哪里？_!_
+6552001447708328206_!_115_!_news_agriculture_!_农村俗语“出门看天色,进门看脸色”是什么意思？_!_
+6552013954900558350_!_115_!_news_agriculture_!_肥料造假新动向：几近乱真！_!_假冒伪劣产品,傍名牌,肥料,肥料造假,造假
+6552367597923336712_!_115_!_news_agriculture_!_江西职业农民带薪种田_!_江西,江西省,永丰县,家庭农场,七都乡
+6552308263562510861_!_115_!_news_agriculture_!_供电公司和农业银行共同牵头实施户用光伏_!_合作社,贫困户,农业银行,村集体,科左后旗
+6552138807414620679_!_115_!_news_agriculture_!_五月车厘子与你相约，美味大樱桃，采摘节开始啦！_!_礼记,大樱桃,樱桃,营养元素,血红蛋白,含桃
+6552275564932104708_!_115_!_news_agriculture_!_人民日报：有些贫困户畏难情绪重 需补精神之“钙”_!_顺势而为,羊肚菌,竹荪菌,贫困群众,贫困地区
+6552299172299538957_!_115_!_news_agriculture_!_蛋鸡养殖效益指标计算_!_回收期,产蛋高峰期,蛋鸡,预混料,产蛋率
+6552358274241396999_!_115_!_news_agriculture_!_柑橘根系总出问题是什么原因？_!_
+6552415177675899150_!_115_!_news_agriculture_!_如何对付柑橘红蜘蛛？_!_
+6551897925436309763_!_115_!_news_agriculture_!_农村地里的野草需要除尽吗？长草会有什么害处？_!_
+6552298146762850830_!_115_!_news_agriculture_!_剪除病叶、卷叶究竟为啥？为什么连杂草都要清理_!_红蜘蛛,菌丝体,病虫害,毒死蜱,介壳虫
+6552319060950909453_!_115_!_news_agriculture_!_这种树木你认识吗？农村的山坡上到处都是，它是一种名贵木材_!_不知春,中药材,农村,黄檀,黄檀木
+6552458441049244168_!_115_!_news_agriculture_!_西藏波密：千户贫困户获“脱贫保”_!_晋美多吉,波密县,脱贫保,批援藏工作队波密工作组,贫困户
+6552344839692747278_!_115_!_news_agriculture_!_四川华蓥：荒山荒坡结出“金蛋蛋”_!_重庆,阳和镇,四川省华蓥市永兴镇陈家湾村,枇杷,金蛋蛋
+6552256843886887428_!_115_!_news_agriculture_!_「脱贫攻坚」贵阳市信访局积极推进脱贫攻坚“春风行动”_!_宅吉乡,堰塘村,信访局,开阳县,贵阳市
+6552490591026938381_!_115_!_news_agriculture_!_吃红心火龙果过时了！2018流行吃黄龙火龙果，1斤才300块_!_猕猴桃,巨峰葡萄,南方北方,核桃苗,百香果
+6552336878421410056_!_115_!_news_agriculture_!_种地不用化肥会怎样？_!_
+6552483903918047748_!_115_!_news_agriculture_!_在农村里，有一种树被称为“蛋树”，你见过蛋长在树上吗？_!_金银茄,蛋树,植物,茄子,农村
+6552299723133288973_!_101_!_news_culture_!_李逵临终前，含泪说了22个字，揭开了宋江的真实面目_!_水浒传,梁山,晁盖,郓城县,宋江,李逵
+6552296537622315534_!_101_!_news_culture_!_山西：木乃伊、狮身人、沙比提 古埃及文明神秘之旅等你来！_!_太阳神,山西博物院,亡灵书,古埃及人,木乃伊,纸莎草
+6552362175866863886_!_101_!_news_culture_!_上联：乱石穿空今夜雨，如何对下联？_!_
+6552482862593999367_!_101_!_news_culture_!_这是字吗？简直是一件艺术品，看着就很舒服，真让人羡慕_!_行书,字帖,硬笔,练字,行楷
+6552484405028323591_!_101_!_news_culture_!_马克思是世界上最伟大的哲学家吗？_!_
+6552425247197364744_!_101_!_news_culture_!_上联：握手一长叹；求下联_!_金井阑,长相思,生民涂炭
+6552241402208583943_!_101_!_news_culture_!_上联：对联，关耳何事，怎么对下联？_!_
+6552366652996976900_!_101_!_news_culture_!_故宫断虹桥那只“护档狮”有什么来历？_!_
+6552369885622567427_!_101_!_news_culture_!_董寿平｜海到无边天是岸，身登绝顶我为峰_!_哲学,内经,大渡河,董寿平,人字瀑,黄山,书法,哲学观点看艺术,书画语录,笔墨,意境,枕上,脱俗章·十六,庄子·养生主,雨后黄山,董老,黄山云海,对立统一
+6552448310722953732_!_101_!_news_culture_!_十字坡一战，孙二娘彻底被武松征服，从此对武松刮目相看_!_孙二娘,水浒传,西门庆,武松,十字坡
+6552281374642929928_!_101_!_news_culture_!_民间的传说，“酸儿辣女”是真的吗？_!_
+6552339233997062403_!_101_!_news_culture_!_玉笛吹来美人风，如何赐句？_!_
+6552376309987672334_!_101_!_news_culture_!_《水浒传》给我们展示了一个怎样的世界和怎样的世界观？_!_
+6552435799462248712_!_101_!_news_culture_!_艺术在进化过程中有什么用？_!_
+6552374827464786189_!_101_!_news_culture_!_没文化，没技术，如何找出路？_!_
+6552221604431003918_!_101_!_news_culture_!_上联：东方有新日，扫却万里愁云，如何对下联？_!_
+6552288090663158024_!_101_!_news_culture_!_上联：青山润色三春梦，怎么对下联？_!_
+6552410551543136515_!_101_!_news_culture_!_上联:青丝一任凌风乱，如何对下联？_!_
+6552315596443222286_!_101_!_news_culture_!_上联：半床明月增寂寞，下联怎么对？_!_
+6552438479781888270_!_101_!_news_culture_!_上联：天高任鸟飞，下联如何对？_!_
+6552283991385309710_!_102_!_news_entertainment_!_明星吃相大比拼，谁最可爱？谁最丑？_!_萌萌哒,赵薇,男子汉,真人秀,张艺兴
+6552354260288274957_!_102_!_news_entertainment_!_继乔任梁后这位明星也不堪重负精神崩溃深夜发文求粉丝放过自己_!_艺人,乔任梁,粉丝,唐艺昕,九州,张若昀
+6552315964505981444_!_102_!_news_entertainment_!_刘亦菲撞衫赵丽颖谁更美？细数明星撞衫大集合！_!_刘亦菲,李小冉,赵丽颖,赵薇,撞衫
+6552390850876277252_!_102_!_news_entertainment_!_张震低调现身机场颜值20年没变 网友：羡慕航空姐姐_!_春光乍洩,主竞赛单元,张震,坎城影展,坎城,牯岭街少年杀人事件,王家卫
+6552411935520850445_!_102_!_news_entertainment_!_明明是大妈年龄，却非得演少女，看到最后一位只想说：还我男神！_!_倾世皇妃,西门无恨,伊能静,同龄人,长歌行,胡歌,焦恩俊,杨钧钧,许晴
+6552495786712629763_!_102_!_news_entertainment_!_《创造101》口碑不敌《偶练》，槽点满满，三期下来你记得谁？_!_偶像,练习生,粉丝,偶像练习生,创造101
+6552361301442560526_!_102_!_news_entertainment_!_180506 恭喜熊梓淇微博粉丝数突破900W大关！_!_粉丝,小浣熊,微博粉丝,熊梓淇
+6552383820681707779_!_102_!_news_entertainment_!_赵丽颖为什么火？_!_
+6552251591859110413_!_102_!_news_entertainment_!_到底是地球人的自我沦丧？还是外星文明的无耻入侵？_!_女子敢死队,外星文明,灯塔,三体,莉娜,湮灭,凯恩
+6552831618497643016_!_102_!_news_entertainment_!_在节目里懒得令何炅都崩溃的李诞，有个特别美貌的女朋友……_!_彭昱畅,刘宪华,张绍刚,李诞,何炅,吐槽大会,今晚80后脱口秀
+6552392255351554567_!_102_!_news_entertainment_!_180506 奶凶不是唯一标签 品味白侦探推理魅力爆发瞬间_!_白侦探,我是大侦探,白敬亭
+6552357455953658371_!_102_!_news_entertainment_!_蔡依林被“完美女神”提问质疑已经过气，她的心态赢得全场喝彩_!_蔡依林,jolin,Jolin,艺人,女神
+6552271643337556238_!_102_!_news_entertainment_!_何为人格魅力？_!_
+6552159269754503688_!_102_!_news_entertainment_!_击败林志玲成为台湾第一美女，如今婚姻幸福让人羡慕！_!_姚元浩,Tony,犀利人妻,隋棠,爱情魔法师,痞子英雄,模特,温尼伯,动物系恋人啊
+6552252500785758471_!_102_!_news_entertainment_!_如果把明星整成网红会是什么样？_!_
+6552386981580505614_!_102_!_news_entertainment_!_拉美裔女明星吉娜·罗德里格兹，街头拍电影亲吻_!_吉娜,罗德里格兹,吉娜罗德里格斯,东方IC,处女孕事,罗德里格斯
+6552145843435078148_!_102_!_news_entertainment_!_孙艺珍也救不了《经常请吃饭的漂亮姐姐》的剧情，已打算弃剧？_!_漂亮姐姐,凉凉,丁海寅,孙艺珍,苏志燮,小狼狗
+6552230928561734147_!_102_!_news_entertainment_!_泰国男明星里的同志“天菜”，快来看泰国小哥哥了！_!_偷心俏冤家,暹罗之恋,叻丕府,恋上黑涩会,逐爱天涯,Push,命定之爱,同志电影,Weir
+6552388595632570883_!_102_!_news_entertainment_!_宋祖儿现身活动，着小公主的打扮也是好可爱呢，笑起来很有感染力_!_宋祖儿,小公主
+6552468663495033351_!_102_!_news_entertainment_!_孙坚送花，潘老师仿佛喜得贵子，来搞笑吗？_!_粉丝,白夜追凶,潘粤明,孙坚
+6552313472103743758_!_103_!_news_sports_!_山东鲁能有没有可能拿到今年的中超冠军？_!_
+6552334517527380228_!_103_!_news_sports_!_世乒赛决赛，刘诗雯2:3不敌伊藤美诚，对此你怎么看？_!_
+6552391015871807757_!_103_!_news_sports_!_为何看到2018NBA季后赛勇士VS鹈鹕G3，勇士球迷称勇士是在放水？_!_
+6552357947052130824_!_103_!_news_sports_!_WWE巨星究竟多有钱？约翰塞纳有私人车库，比WWE老板的车还多！_!_奥斯汀,罗林斯,约翰塞纳,WWE,米兹
+6552424162382577924_!_103_!_news_sports_!_大连一方主场3-0击败广州恒大，取得中超首胜，他们取得胜利的原因是什么？_!_
+6552265691829895428_!_103_!_news_sports_!_为什么国人能接受并喜欢石川佳纯，但不怎么喜欢张本智和？_!_
+6552827016679260676_!_103_!_news_sports_!_日媒：日乒和中国的差距正在缩小_!_朱雨玲,平野美宇,哈尔姆斯塔德,伊藤美诚,石川佳纯
+6552130007089545479_!_103_!_news_sports_!_上港1-2输给国安的比赛中，国安后卫刘欢的手球，没有判罚点球，是不是误判？_!_
+6552420167714865416_!_103_!_news_sports_!_网传《马卡报》披露，内马尔希望加盟皇马搭档C罗，可能性有多大？_!_
+6552468772169449987_!_103_!_news_sports_!_德甲官方第33轮最佳阵容：J罗入选 马里奥-戈麦斯在列_!_莱比锡,美因茨,斯图加特,沙尔克04,最佳阵容
+6552322629083070984_!_103_!_news_sports_!_凯西教练：我们这个赛季已经命悬一线_!_凯西,主教练,客场
+6552268006032933127_!_103_!_news_sports_!_武当掌门游玄德在生活中。比如街斗对阵泰拳王西提猜，结果会怎样？_!_
+6552463785402040836_!_103_!_news_sports_!_猛龙惨败骑士真因曝光！两错误让詹姆斯上演绝杀，这锅凯西得背！_!_主教练,猛龙,德罗赞,詹姆斯,骑士
+6552396158507942404_!_103_!_news_sports_!_鲁能0-0客平人和，错失登顶良机_!_刘彬彬,王大雷,金敬道,吴兴涵,U23
+6552256995334816269_!_103_!_news_sports_!_女排国家联赛已成新人试金石，她若能把握住上位良机必成大器！_!_李盈莹,刘晓彤,北京女排,土耳其,张常宁
+6552417155684499725_!_103_!_news_sports_!_如果本赛季詹姆斯率领骑士夺冠，詹姆斯将有怎样的历史地位？_!_
+6552431583494668803_!_103_!_news_sports_!_这一球之后，来看看各路球星怎么就变“詹吹”了_!_泰伦卢,麦科勒姆,布拉德利,勒布朗,詹姆斯
+6552395769075204612_!_103_!_news_sports_!_爵爷，请你为自己再赢一次！_!_索尔斯克亚,吉格斯,爵爷,曼联,维迪奇
+6552131700271350279_!_103_!_news_sports_!_刘诗雯首负日本因策略不当？8年磨砺自我救赎，刘国梁赞她这一点_!_刘国梁,朱雨玲,刘诗雯,平野美宇,丁宁
+6552421236192838148_!_104_!_news_finance_!_关于配额、新疆棉纺等话题，中棉行协第五届六次常务理事会作热议_!_新疆,短纤维,郑州商品交易所,棉花,期货
+6552407376924770829_!_104_!_news_finance_!_上交所高层变动！总经理黄红元将升任理事长_!_理事长,中国人民银行,黄红元,上海证券交易所,理事会
+6552175311398633997_!_104_!_news_finance_!_铭硕言5.6黄金非农急催拉盘空单扫尾，亏损该怎么对待后市_!_BOLL,黄金,日均线,MACD,MA5
+6552198073928384776_!_104_!_news_finance_!_位置很差的水果店，如何通过一招将生意做成社区内最好？_!_
+6552274980187406862_!_104_!_news_finance_!_保险关联交易监管法11年后大修 比现行规定有松有紧_!_自然人,中华人民共和国公司法,办法,保险公司管理规定,公司关联交易管理,公司关联交易,老办法,中华人民共和国保险法,新办法,关联方
+6552272023140172301_!_104_!_news_finance_!_证金公司逆市增持券商股 中小盘成长股逐渐被资金锚定_!_中国人寿,持续,估值,华泰证券,泰格医药,海螺水泥,晶盛机电,红周刊,工商银行,鲁抗医药,表现,银行股,概念股,独角兽,生物医药,康达尔,券商股,供给侧,盈方微,国泰君安
+6552268629390393859_!_104_!_news_finance_!_2018年A股百亿富豪：何享健成首富 贾跃亭退出榜单_!_王卫,周群飞,2018年A股百亿富豪榜,周鸿祎,何享健,百亿
+6552475163718320397_!_104_!_news_finance_!_白马蓝筹股的价格突然下跌，原因有哪些？_!_
+6552440183998906884_!_104_!_news_finance_!_中国股市的“庄家”到底是谁_!_股市,平均线,母猪,炒股,股价,日均线,均线,股票,庄家,解套
+6552463343234318852_!_104_!_news_finance_!_龙头惨遭砸盘透重大暗示 下周初变盘风暴或骤然来袭？_!_星星科技,股市,通化东宝,投资者,a股
+6552429418151674372_!_104_!_news_finance_!_公告精选：药明康德将于5月8日上市；南山控股换股吸收合并深基地B_!_超过,新能源汽车,晚公告,药明康德,华信,鸿利光电,南山控股,力帆股份,新能源
+6552306167220011533_!_104_!_news_finance_!_这类股你敢买吗？_!_a股,伯克希尔,股市
+6552377101167624707_!_104_!_news_finance_!_海富通FOF镀金未果 仓促建仓择时失败业绩陷泥潭_!_FOF,基金经理,巴克莱,海富通基金,海富通,建仓期
+6552474357485011208_!_104_!_news_finance_!_最近股票亏了4万多，本金只有10万，还有必要坚持吗？_!_
+6552289128770175239_!_104_!_news_finance_!_自3月23日以来，A股持续在3100点附近震荡，股市何时能突破该区域？_!_
+6552395076885021192_!_104_!_news_finance_!_币圈重磅：BTC牛市已来，教你如何把握抄底时机！_!_处于,抄底,生命线,底部,长期
+6552367816870199821_!_104_!_news_finance_!_中国中药协会灵芝专业委员会成立大会_!_灵芝,中国中药,中国中药协会,保健品,寿仙谷,浙江省
+6552457567044370692_!_104_!_news_finance_!_增值税下调后，小规模纳税人开票如何处理？_!_
+6552355047202619918_!_104_!_news_finance_!_今日消息面：BCH上涨近3倍，高晓松称数字货币可理解为地下黄金_!_TRX,数字货币,区块链,阿塞拜疆,CNBC,Colas,比特币
+6552382385961304589_!_104_!_news_finance_!_紫鑫药业董事长郭春林：三大业务为核心 迎接大健康时代来临_!_郭春林,基因测序仪,紫鑫药业,人参,OTC
+6552284733064086030_!_101_!_news_culture_!_“福”字倒贴的由来_!_马大脚,走马灯,太祖,马皇后,恭亲王府
+6552402632751710467_!_101_!_news_culture_!_云横东岭松挺立；有什么好的下联呢？_!_
+6552276708576198920_!_101_!_news_culture_!_上联：秋水映月一船梦。如何对下联？_!_
+6552408714211492100_!_101_!_news_culture_!_上联：无可奈何花落去，下联怎么对？_!_
+6552363730959598093_!_101_!_news_culture_!_京剧油画图集/_!_油画,画室界黄埔军校孪生,黄埔军校,画室,京剧
+6552494533995659533_!_101_!_news_culture_!_古代对女子的称呼有哪些？_!_
+6552241795143565576_!_101_!_news_culture_!_小学美术教师编好的书该怎样出版？_!_
+6552297281859617032_!_101_!_news_culture_!_“轻风细柳，淡月梅花”，在这两句诗中各加一个字作腰，加哪两个字最好？_!_
+6552428094622269955_!_101_!_news_culture_!_「诗歌」梦里_!_五河县,诗歌,张新文,江苏省
+6552274085156487432_!_101_!_news_culture_!_“茶叶女”是如何行骗的？利用了人们什么心理？_!_
+6552457148482191620_!_101_!_news_culture_!_上句：麻将桌上观君子，如何接下句？_!_
+6552287473689428231_!_101_!_news_culture_!_墨汁里可以加酒吗？_!_
+6552423574630564099_!_101_!_news_culture_!_上联：九日为旭，人人从，下联是什么？_!_
+6552252998435733768_!_101_!_news_culture_!_青龙戏白虎，江涛乐夜风。如何接句？_!_
+6552409795364651267_!_101_!_news_culture_!_春风欲走花言谢，下联如何对？_!_
+6552191189133558030_!_101_!_news_culture_!_上联：江山千古秀，怎么对下联？_!_
+6552240220698313220_!_101_!_news_culture_!_七律·洛阳牡丹_!_洛阳牡丹,七律
+6552442441599812103_!_101_!_news_culture_!_李白、杜甫、陈子昂三人写白帝城的诗，看其心境与悲欢_!_陈子昂,巴子国,杜甫,白帝,下江陵,李白
+6552224248625103107_!_101_!_news_culture_!_上联：北京北，南京南，如何对下联？_!_
+6552406507948868110_!_102_!_news_entertainment_!_胡彦斌和女选手yamy撞脸俞灏明：这是不是你妹妹？_!_小笼包,胡彦斌,中国有嘻哈,云开见月明,陈妍希,俞灏明
+6552339549173842445_!_102_!_news_entertainment_!_「EXO」「新闻」180506 剧情发展让人猝不及防 李代表那晚到底做了什么？_!_济州岛,exo,iHQ,Richman,richman,Rich Man
+6552458475140547085_!_102_!_news_entertainment_!_《后来》“退票门”或是“罗生门”，猫眼真的会自毁前途？_!_猫眼,前任3,罗生门,捉妖记2,五一档,叶问3,刷票房,票务平台
+6552426759277511181_!_102_!_news_entertainment_!_蔡徐坤粉丝见面会被激光笔照射 粉丝们你们要冷静啊！_!_激光笔,Percent,蔡徐坤,偶像练习生,Nine
+6552348162860777997_!_102_!_news_entertainment_!_娱乐圈新晋小鲜肉大盘点，各个才艺双全，来看看有你的爱豆吗？_!_蔡徐坤,何猷君,跑男,Percent,最强大脑,Nine,偶像练习生,范丞丞
+6552287847909425677_!_102_!_news_entertainment_!_毛晓彤出席活动，网友：一条朴素的连衣裙和她甜美的笑容真耐看！_!_连衣裙,笑容真耐看,毛晓彤,毛晓彤出席活动
+6552387097091637764_!_102_!_news_entertainment_!_“零差评”的4部网剧，安悦溪上榜2部，最后一部总播放量破40亿！_!_八王爷,汪东城,安悦溪,郑业成,镇魂街
+6552268586793042435_!_102_!_news_entertainment_!_蒋欣出席活动，网友：这个发型很配这套礼服！_!_蒋欣,发型,礼服,蒋欣出席活动
+6552462721021903373_!_102_!_news_entertainment_!_心疼蔡徐坤被激光伤害，嫌疑人声称就是要照死他，彰显人性！_!_演唱会,蔡徐坤
+6552293057897693704_!_102_!_news_entertainment_!_史上最牛的一部影视作品：很多人都没有看过生存之民工值得欣赏_!_管虎,民工,孙松,陶泽如,渴望,生存之,生存之民工
+6552369117469344270_!_102_!_news_entertainment_!_71岁施瓦辛格露面，跟小女友穿情侣装，心脏手术后走路迟缓_!_施瓦辛格,情侣装,电影
+6552473695703532036_!_102_!_news_entertainment_!_张一山生日杨紫画画送祝福，网友提问直击灵魂：为什么都穿小裙子_!_家有儿女,柒个我,杨紫,张一山,手牵手
+6552277942846620167_!_102_!_news_entertainment_!_陈冲出席活动，网友：脖子上的项链非常精美！_!_陈冲,脖子,陈冲出席活动,项链
+6552393738583278088_!_102_!_news_entertainment_!_《跑男》请谁也不会请她们！这三位女星，其中有位请来会太尴尬！_!_奔跑吧兄弟,刘亦菲,女星,黄晓明,综艺节目,李菲儿
+6552356149708980749_!_102_!_news_entertainment_!_白富美代表綦美合为了陈飞宇手撕欧阳娜娜？网友：王文也没参与吗_!_綦美合,美合,王文,秘果,欧阳娜娜,陈飞宇
+6552296098805842445_!_102_!_news_entertainment_!_王俊凯竟要挑战这么惊险的人生第一次！高能量体质体验，勇攀高峰_!_王俊凯,此时此刻,高能少年团
+6552322521180406286_!_102_!_news_entertainment_!_古装剧中那些男扮女装，第一个认不出来，最后一个至今无人超越_!_古装剧,男扮女装,怪侠一支梅,红楼梦,马天宇,陆毅,女扮男装
+6552439093542781454_!_102_!_news_entertainment_!_《爸爸去哪儿6》嘉宾已经确认阵容强大颜值超高，网友：还有吴京？_!_爸爸去哪儿第六季,爸爸去哪儿6,纪凌尘,妈妈是超人,张丹峰,爸爸去哪儿,包贝尔,王力宏,张晞彤
+6552353507666559492_!_103_!_news_sports_!_詹姆斯38分压哨抛投绝杀 骑士3-0猛龙拿到赛点_!_瓦兰丘纳斯,猛龙,詹姆斯,骑士,德罗赞
+6552450513877598733_!_103_!_news_sports_!_国脚张呈栋：从没想过自己会出一本书_!_尹鸿博,职业足球运动员,张呈栋,足球,这本书
+6552294007513284868_!_103_!_news_sports_!_离开新疆的李秋平，下赛季会到哪个球队执教？_!_
+6552457929553871373_!_103_!_news_sports_!_过人如本泽马？卡拉斯科点赞后引来马竞球迷指责_!_过人,卡拉斯科,本泽马,广州恒大,马竞球迷
+6552314034375361027_!_103_!_news_sports_!_中国自己的“球王”：22岁新星戏耍恒大3国脚太强悍！_!_李学鹏,恒大,大连一方,于汉超,汪晋贤
+6552299096776900878_!_103_!_news_sports_!_曼联传奇主帅弗格森突发脑溢血入院，现在情况如何？_!_
+6552275557508186632_!_103_!_news_sports_!_侯森：给自己定位代班门将，等杨智复出国安会更进一步_!_足球,侯森,国安会,门将,胡靖航
+6552303032694997517_!_103_!_news_sports_!_易建联接班人晒邀请函 秀警句激励自己不忘初心_!_中国男篮,邀请函,董瀚麟,易建联,国家队
+6552271555315892744_!_103_!_news_sports_!_西蒙斯：我们凉凉了，0-3没希望了，球迷：你们需要詹姆斯，追吧_!_西蒙斯,詹姆斯,季后赛,凯尔特人,76人
+6552393718710665480_!_103_!_news_sports_!_丢一场赢一场，今天刘诗雯在世乒赛女团决赛的表现值不值得表扬？_!_
+6552315907429892356_!_103_!_news_sports_!_跑马拉松你是如何准备的，要准备些什么？_!_
+6552339767634166285_!_103_!_news_sports_!_NBA和CBA差距在哪里？6张图一目了然！_!_NBA,朱芳雨,cba,雷阿伦,CBA
+6552364293180883213_!_103_!_news_sports_!_上港主场1-2负于国安，遭遇联赛两连败，上港到底输在哪？_!_
+6552368220118974728_!_103_!_news_sports_!_骑士VS猛龙G3，乐福在内线强硬一打三，你怎么看？_!_
+6552251002148356615_!_103_!_news_sports_!_竞篮2串1：勇士客场擒鹈鹕 爵士主场再负火箭_!_米切尔,新奥尔良,勇士,季后赛,爵士
+6552421630524523015_!_103_!_news_sports_!_骑士官宣一消息，詹姆斯做出回应，泰伦卢成最大赢家_!_骑士,泰伦卢,詹姆斯,猛龙,欧文
+6552147950871511565_!_103_!_news_sports_!_皇马2大铁主力将无缘国家德比_!_伊斯科,皇马,国家德比,马卡报,拜仁
+6552468760572199428_!_103_!_news_sports_!_这支球队老板出场踢三场足协杯创历史，球队三轮狂丢14球也创纪录_!_中国足球,延边北国,金学建,足协杯,中乙联赛
+6552235119124414980_!_103_!_news_sports_!_世界残疾人射击锦标赛，河北姑娘破世界纪录夺冠_!_河北,残疾人,锦标赛,CCTV5,SH1
+6552476765137142023_!_103_!_news_sports_!_武当掌门游玄德在生活中。比如街斗对阵泰拳王西提猜，结果会怎样？_!_
+6552368507911143949_!_104_!_news_finance_!_全国人民喜迎油价上涨！_!_石油,原油,中石化,CNBC,OPEC,欧佩克,油价
+6552252212754514183_!_104_!_news_finance_!_如果你已然衣食无忧，有200万闲钱做投资，你会如何使用？_!_
+6552451014669107726_!_104_!_news_finance_!_你还在为企业和项目的资金而发愁嘛？你会融资嘛？懂得如何融资嘛_!_投资者,天使投资,所有者,周志轩,张启明
+6552488950181658884_!_104_!_news_finance_!_余额宝升级以后，预期收益率是否会回到6%？_!_
+6552312174981677319_!_104_!_news_finance_!_余额宝升级了，有什么不一样？_!_
+6552431925200421380_!_104_!_news_finance_!_Brian Kelly：比特币在短期内将继续激增的三个原因_!_Kelly,加密货币,银行,比特币,美国证券交易委员会
+6552276713038938628_!_104_!_news_finance_!_新妖出世，坐等加冕_!_永和大王
+6552130714656047364_!_104_!_news_finance_!_持有的股票一直不卖能等到牛市大涨吗，什么时候才能牛市大涨？_!_
+6552228437988212996_!_104_!_news_finance_!_5月5日美股低开高走，大涨将近400个点，对下周一A股走势有什么影响？_!_
+6552341028324508163_!_104_!_news_finance_!_盘点全国环保行业最赚钱上市公司50强排行榜_!_重庆水务,龙净环保,中国平安,中国银行,交通银行,农业银行,兴业银行,中国石化,中山公用,首创股份,浦发银行,建设银行,环保行业,工商银行,碧水源
+6552282585152619015_!_104_!_news_finance_!_“股神”巴菲特被一个8岁女孩问倒了！甚至怀疑自己买错股票_!_巴菲特,BNSF,哈撒韦,美国运通,轻资产,伯克希尔
+6552354051940418061_!_104_!_news_finance_!_马克思的影响从未离开——法国再掀马克思主义研究热潮_!_恩格斯,神圣家族,1844年哲学和经济学手稿,法兰西内战,德法年鉴,共产党宣言,巴黎,资本论,资本主义,法国,巴黎手稿,政治经济学批判大纲,巴黎公社运动
+6552374496534200584_!_104_!_news_finance_!_欠钱不还的人是什么心态？有钱也不愿意还是什么心态？_!_
+6552476841590915597_!_104_!_news_finance_!_450亿！又一500强民企跨了 释放哪些信号_!_盾安,中国民营企业500强,浙商银行,民营企业,现金流
+6552346254494401037_!_104_!_news_finance_!_“移动医疗第一股”平安好医生港交所挂牌上市，融资11亿美元_!_医疗,贝莱德,互联网医疗,平安,BlackRock,健康管理,上市
+6552454246028018179_!_104_!_news_finance_!_提供借款后立即要求支付利息属于 变相预先扣除利息_!_借款人,借款,出借人,利息
+6552333584743531012_!_104_!_news_finance_!_下周核心股票池：世界杯要来，最强最火的居然是只鸭脖概念股_!_价值型资金流向,老板电器,波段型,持有型,概念股,净流入额
+6552127585629765901_!_104_!_news_finance_!_信用卡如何养卡，如何提额快？_!_
+6552374235614937614_!_106_!_news_house_!_海口绿地海长流开发商状告业主传播谣言 业主拿出质量报告反击_!_工程质量,海长流,民事起诉状,补充协议,现浇板,海口市商品房买卖合同,商品房买卖合同,海口
+6552394602094002435_!_106_!_news_house_!_西安房价会涨过2万的均价么，如果跌的话还能跌回原来的6千么？_!_
+6552474032057352452_!_106_!_news_house_!_发布落户宽松与人才引进政策，天津、青岛加入“抢人”大战。你觉得这些城市对年轻人有吸引力吗？_!_
+6552298933748498957_!_106_!_news_house_!_深圳布吉哪些地方有旧改拆迁的项目？_!_文博宫,求水山公园,深圳,华侨城,布吉
+6552291797484175886_!_106_!_news_house_!_人人都承认房价存在泡沫，但为什么没人敢说房价下跌？_!_房价,时间点,房地产,任志强,谢国忠
+6552275845266800900_!_106_!_news_house_!_2018年随着二胎政策放开，三四线城市房价还在上涨，这两者有何关系，房价上涨的原因有哪些？_!_
+6552410388078526989_!_106_!_news_house_!_房产税最新消息，专家：与房价扯不上一毛钱的关系，别等了_!_地产商,房地产,任志强,房产税,樊纲
+6552462842195345924_!_106_!_news_house_!_浩大的都江堰万达文旅城分商业和住宅两部分，住宅转由融创开发了_!_万达集团,万达城,赵公山,都江堰,万达
+6552083834505003527_!_106_!_news_house_!_房价从3万跌到1万，这绝对是个例！_!_房价,燕郊,炒房者,购房者,外地人
+6552482548784562445_!_106_!_news_house_!_买房付完首付后要到一个月才能拿到购房合同正常么？_!_
+6552401073976377870_!_106_!_news_house_!_上海4月份住宅供应涨84.5%，成交量创8年来新低_!_上海,购房者,商品住宅,新增供应,新建商品住宅
+6552326552154014211_!_106_!_news_house_!_名为房地产合作开发实为用地补偿合同的处理_!_国泰公司,最高人民法院,监察局,肖扬,协议书,合作开发合同
+6552304041928425987_!_106_!_news_house_!_大西安2018新开业的41座商业体展示！敢问这是几线城市发展实力？_!_
+6552475582553129223_!_106_!_news_house_!_兰州率先“松绑”限购透出哪些深意和动向？台州又将会如何呢？_!_
+6552491853923483907_!_106_!_news_house_!_如果你在北京有一套房子价值1000万左右，你会不会卖掉房子拿着钱回老家？_!_
+6552465807140454915_!_106_!_news_house_!_业内：东莞升级调控预期较大 房价下降可能性存在_!_三价合一,成交量,五一,房贷利率,东莞
+6552270557037986062_!_106_!_news_house_!_未来张家口房价真的会暴跌吗？_!_
+6552244596506624269_!_107_!_news_car_!_美国成了日本车的天下，为什么美系车在本土迅速堕落？_!_
+6552382303539036676_!_107_!_news_car_!_交警提醒：驾照扣12分和24分有什么区别？这些情况你一定要知道！_!_记分周期,驾驶证,机动车驾驶证,实习期,机动车驾驶人
+6552296323180134915_!_107_!_news_car_!_想要干爽迎接夏天？这个汽车SPA少不了_!_SPA,汽车内饰,空调系统
+6552371654608027907_!_107_!_news_car_!_长时间震颤会怎么样？_!_
+6552358697899655684_!_107_!_news_car_!_幸福多一种选择——宋MAX 6座华中区星级上市 售价7.99万元起_!_幸福加推,星级,比亚迪,比亚迪汽车,比亚迪宋MAX 6,华中区
+6552413900166398471_!_107_!_news_car_!_全新DS7起步不到21万，1.6T+8AT，远超GLA宝马X1_!_DS 7,雪铁龙,Nappa,EMP2,DS品牌,DS,法国人
+6552461192072265988_!_107_!_news_car_!_机油滤芯有一次性和可换纸滤芯的，为什么有设计区别？_!_
+6552450692852744707_!_107_!_news_car_!_大货车的上的90后美女司机，行程几十万令人佩服！_!_司机,货车,石家庄,美女
+6552474004832125187_!_107_!_news_car_!_一汽大众速腾1.6L手动家用动力如何？_!_
+6552136234922148356_!_107_!_news_car_!_长安新车撞脸大众，比宝骏510大一圈只要6万起，又将是爆款？_!_宝骏,CS75,SUV市场,CS35,宝骏510,现款车型
+6552432350951637511_!_107_!_news_car_!_大货车司机讲大实话：“为什么我们不踩刹车”！_!_小轿车,小汽车,货车,远光灯,货车司机
+6546444089301139975_!_107_!_news_car_!_一款最适合群众的房车｜实拍金杯海狮C型房车_!_坐沙发,抽油烟机,电磁炉,房车,金杯海狮
+6552205319668236807_!_107_!_news_car_!_夏季开车，这几种东西老司机都不会放在车上_!_打火机,电池,车用香水,小家电,放大镜
+6552493571629384195_!_107_!_news_car_!_「人人车」女汉子买16款路虎-揽胜极光，30万入手，霸气测漏吧_!_二手车市场,二手车,路虎,揽胜极光
+6552263376959963655_!_107_!_news_car_!_宝马第一款三排SUV：舒适性和交互性之间达到了极好的平衡_!_雷克萨斯,SUV,BMW,宝马,X7,X5
+6552412529820172548_!_107_!_news_car_!_为什么SUV没有轿车舒服，买的人还那么多？_!_
+6552452756840382728_!_107_!_news_car_!_众泰T3001.5手动豪华落地多少钱？_!_
+6552093650476597773_!_107_!_news_car_!_丰田在华想翻身就靠它了！比宝马5系还美，卖22万现在就存钱！_!_丰田,宝马5系,新皇冠,皇冠
+6551198935707288068_!_107_!_news_car_!_国产车推出高端线路品牌，性价比更高，车主反馈不错_!_国产汽车,合资车,进口车,汽车市场,国产车
+6552319370985472264_!_108_!_news_edu_!_三本的学生一定会比211、985混的差吗？_!_
+6552260186885587464_!_108_!_news_edu_!_高考，能上重点大学，可是因触犯底线，他竟然跪下了_!_数学科,高考,河南省,社旗县,监考老师
+6552443024163471885_!_108_!_news_edu_!_敢于质疑是高等教育的价值所在——引自新书《不要假装读大学》_!_质疑,不要假装读大学,知识,洪秀全,独立思考,批判性思维
+6551760066372960782_!_108_!_news_edu_!_惊心炫目，小学还能开三语班？昆明仅昆理工附小一家_!_高年级,昆明理工大学,附属中学,DSD,留学
+6552270375059718659_!_108_!_news_edu_!_两所民办高等非学历教育机构不合格_!_合格学校,主管部门,普通中学,民办学校,职业中学
+6552279442247385351_!_108_!_news_edu_!_面对考场作文中出现的一些突发状况，究竟该如何应对？_!_
+6552341161397191176_!_108_!_news_edu_!_孩子真的在幼儿园被孤立？家长先反思，再听老师怎么说_!_小朋友,益智玩具,幼儿园
+6552342157636665869_!_108_!_news_edu_!_温暖龙岩之龙院前身一_!_龙岩,龙岩师范高等专科学校,教职工,大专班,师范专科学校
+6552489224325562632_!_108_!_news_edu_!_山西考生超一本130分，能上什么大学？_!_
+6552396177994678791_!_108_!_news_edu_!_南京：对欺凌学生等行为零容忍 从重从快依法惩处_!_校园欺凌,禁止设立,南京,性侵害,教师
+6552395296452641028_!_108_!_news_edu_!_九年义务教育结束后，上职业技术学校好，还是技校好？_!_
+6552370199947903245_!_108_!_news_edu_!_把孩子送去夏令营合适吗？_!_
+6552445101719683332_!_108_!_news_edu_!_为什么江苏大学不在南京？_!_
+6551976783229485571_!_108_!_news_edu_!_美术生去北京画室参加集训会不会影响联考成绩？_!_北京,美术生,省统考
+6552365705654698503_!_108_!_news_edu_!_百年五四堪回首──关于五四运动的前前后后_!_辛丑合约,林长民,飞毛腿,北洋政府,可怜之人力车夫,巴黎和会,骆驼祥子,徐世昌,解决山东悬案细目协定,人力车夫,薄奠,新青年,洋车夫,梁启超,汪大燮,一件小事,车毯
+6552308677263491342_!_108_!_news_edu_!_西安小升初电脑摇号加面试，要不要报面试培训班？_!_
+6552257300608844291_!_108_!_news_edu_!_让孩子赢在起跑线上_!_儿童英语,学习英语
+6552311676513812744_!_108_!_news_edu_!_东财的会计专硕在财经院校里水平怎么样？_!_
+6552268850086281741_!_108_!_news_edu_!_2018初级会计职称考试判断题怎样做才能不被倒扣分？新北会计培训_!_限定词,判断题,初级会计,会计,会计职称
+6552249963433165326_!_108_!_news_edu_!_毕业季来临台生“登鹭”实习忙 厦门青创基地积极“搭台”_!_厦门,大陆,台湾,104人力银行,张培轩
+6552294945116389901_!_109_!_news_tech_!_小米7未到！这两款小米手机目前性价比最高，米粉：可惜买不到_!_AI,小米MIX,雷军,小米手机,OPPO
+6552397963136598535_!_109_!_news_tech_!_5月5日她味红糖亮相京东众筹首页_!_京东,味红糖
+6552456954332053768_!_109_!_news_tech_!_有什么高逼格的微信名字？_!_
+6552309988532945166_!_109_!_news_tech_!_怎样做一个类似拼多多的团购+分销的APP？_!_
+6552331078948880899_!_109_!_news_tech_!_零基础学开亚马逊之一卖家后台首页讲解_!_模块,搜索框,卖家,亚马逊,fba库存
+6552394585581027598_!_109_!_news_tech_!_1699的魅族m15和1268的魅族pro7标准版该选哪个？_!_
+6552285507370352900_!_109_!_news_tech_!_苹果6 10.5系统有必要升级到最新11的系统吗？_!_
+6552420693768667651_!_109_!_news_tech_!_爆发的微信小程序，电商的又一春？_!_蘑菇街,微信,马化腾,程序,开普勒
+6551662172072575492_!_109_!_news_tech_!_PS利用滤镜制作立体文字_!_滤镜
+6552463741403791629_!_109_!_news_tech_!_微信滴滴不能用了怎么办？_!_
+6552373182106436109_!_109_!_news_tech_!_估值近千亿美元！小米凭啥值这么多钱？真相了_!_智能手机,小米,小米手机,招股书,互联网服务,互联网公司
+6552405437952557315_!_109_!_news_tech_!_国产手机那么好，为什么还会有这么多人用苹果手机？_!_
+6552282569570779405_!_109_!_news_tech_!_1699的魅族m15和1268的魅族pro7标准版该选哪个？_!_
+6552397074422300942_!_109_!_news_tech_!_为什么大学的程序设计老师不去外面公司做程序员？_!_
+6552479125834039566_!_109_!_news_tech_!_苹果，ipad玩绝地求生怎么录制精彩视频？_!_
+6552481071202566407_!_109_!_news_tech_!_如何上传图片被百度收录？_!_
+6552341505309147399_!_109_!_news_tech_!_老板让我找视频会议系统，有什么好的品牌软件推荐？_!_
+6552117027257975299_!_109_!_news_tech_!_不玩AI套路 360手机N7诠释智能本质_!_Engine,360手机,Hybrid,Transform,AI
+6552141125522555144_!_109_!_news_tech_!_小米max2和红米5p，哪个好一点？_!_
+6552259674001899780_!_109_!_news_tech_!_为什么在我们的观念里，外星人的科技一定会强过人类？_!_
+6552343896658018820_!_110_!_news_military_!_俄罗斯胜利日阅兵总彩排盛况 央视记者带你先睹为快_!_王晓雪,阅兵,卫国战争,阅兵式,俄罗斯
+6552401061242470663_!_110_!_news_military_!_如果发生世界大战，你会怎么办？_!_
+6552279984755441934_!_110_!_news_military_!_什么样的军事条件才算得上是军事大国？_!_
+6552293422806335752_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552450622220665352_!_110_!_news_military_!_装备通用问题破解！新型反舰导弹问世，所有军舰都能用_!_反潜导弹,军舰,反舰导弹,通用性,导弹
+6552421857390232071_!_110_!_news_military_!_10国上门求购2000枚中国鱼雷_!_鱼雷,潜艇,火箭助推鱼雷,助推鱼雷,导弹
+6552365750537945348_!_110_!_news_military_!_往海中投一枚原子弹会发生什么现象？_!_
+6552389030124716296_!_110_!_news_military_!_歼20座舱盖上的两条“花纹”是什么？_!_
+6552280205820428814_!_110_!_news_military_!_二战最好的5大狙击步枪 盘点_!_瞄准镜,手动枪机,春田步枪,98K,狙击步枪
+6552329060893065736_!_110_!_news_military_!_推进44毫米挽救4000人生命，山东小伙把28岁青春融入长空_!_张德山,居民区,冯思广,跳伞,东方IC
+6552284163037200648_!_110_!_news_military_!_南阳籍将军有哪些？_!_
+6552450937137398023_!_110_!_news_military_!_《战狼2》中吴京用铁丝板档火箭弹，这在现实中可能吗？_!_
+6552405143017488643_!_110_!_news_military_!_叙利亚政府军从一处叛军据点缴获的武器里面有俄制的装备，你怎么看？_!_
+6552293029720359171_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552404239367274765_!_110_!_news_military_!_如果马其顿军队遭遇大秦军队，谁会赢得胜利？_!_
+6552299400666808580_!_110_!_news_military_!_同样是国家陷入内战，为何卡扎菲最终丧命而阿萨德一直稳坐江山？_!_
+6552082052211016199_!_110_!_news_military_!_美军A-10攻击机，究竟是如何一战成名的？_!_A-10攻击机,A-10,雷达站,伊拉克,装甲车辆,海湾战争
+6552374476196020750_!_110_!_news_military_!_「原创」酹樽：浅论军队实战化建设之“实”——解放军如何取胜？_!_科幻化,实战化建设,自然系统,实战化,社会系统
+6552487141195120909_!_110_!_news_military_!_为何法国巴士底狱暴动解放的人数竟然只有7个？_!_
+6552469437524476173_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552404225639318020_!_112_!_news_travel_!_千人相亲大会，单身的有福利来了！_!_
+6552474832099868932_!_112_!_news_travel_!_在镇上开一家儿童乐园怎么样？_!_
+6552381703610958084_!_112_!_news_travel_!_为什么唐山有钱人这么多？_!_
+6552244076731695619_!_112_!_news_travel_!_独爱斐济的10个理由！让你彻底徜徉在浩瀚蓝色的天堂度假胜地_!_度蜜月,喜来登,斐济,珊瑚礁,公牛鲨
+6552389808654647821_!_112_!_news_travel_!_如果你是新手，想骑行西藏318国道，选择自行车应避免这几个雷区_!_自行车,美利达,骑行,捷安特,山地车
+6552391857404379662_!_112_!_news_travel_!_国内小众旅游景点，个人感觉还可以，经济实惠的穷游_!_天文台,岬角,南麂岛,麦克马洪线,察隅
+6552296982335980040_!_112_!_news_travel_!_会奖旅游之会议酒店推荐：上海璞丽酒店_!_蒸汽房,上海璞丽酒店
+6552367102680891918_!_112_!_news_travel_!_没有对比就没有伤害，我和你走的好像不是一个川藏线_!_纳木错,扎西半岛,西藏,川藏线,澜沧江
+6552444051981861128_!_112_!_news_travel_!_宁波的天宫庄园适合去秋游吗？_!_
+6552333383362412814_!_112_!_news_travel_!_如何巧妙地给别人留下良好印象？_!_
+6552416314944651783_!_112_!_news_travel_!_宝鸡石鼓园美景_!_宝鸡石鼓园宝鸡石鼓园,中国青铜器,宝鸡石鼓园,石鼓园宝鸡石鼓园,宝鸡石鼓园宝鸡
+6552472362850189581_!_112_!_news_travel_!_你感觉哪个是天津有代表性的美食？_!_
+6552336605095395587_!_112_!_news_travel_!_中国四大园林分别在哪？_!_
+6552393233400332551_!_112_!_news_travel_!_外出旅游时，吃东西被坑的最惨的一次是怎么样的？_!_
+6552465046268543236_!_112_!_news_travel_!_国内哪一家航空公司的飞机餐最好吃？_!_
+6552228181942731011_!_112_!_news_travel_!_石家庄夏季有休闲避暑的景区吗？_!_
+6552497902604780040_!_112_!_news_travel_!_头条号得粉丝得未来_!_
+6552247361979023886_!_112_!_news_travel_!_凤凰旅游，如果不体验这个项目，那真算是白来了_!_黄永玉,沱江泛舟,边城,吊脚楼,万寿宫,万名塔
+6552350577425121795_!_112_!_news_travel_!_胖子游记-大美甘肃张掖篇（二）_!_大般若经,河西走廊,甘州府志,寺院,大佛寺,张掖,木塔
+6552143031384932616_!_113_!_news_world_!_特朗普只是一个商人，更是一个政治素人，美国如此发达的国家为什么会选其当总统？_!_
+6552468012232868110_!_113_!_news_world_!_美国建国不过两百年，为什么会成为世界强国？_!_
+6552247676983837198_!_113_!_news_world_!_缅甸帕敢矿区再发山体滑坡 已致至少12人死亡_!_克钦邦,遇难者,缅甸,山体滑坡,帕敢
+6552461663600116237_!_113_!_news_world_!_期限或变成无限！为何敢于对美说不？手握杀手锏和美软肋！_!_印度洋,伊朗,发展核武器,俄罗斯,叙利亚
+6552384877507904013_!_113_!_news_world_!_这些真实而又残酷的历史照片_!_世贸中心,难民营,纽约,阿富汗,伊拉克
+6552385845385495054_!_113_!_news_world_!_如何与特朗普打交道？外交官：要夸他比奥巴马强_!_默克尔,外交官,特朗普,内塔尼亚胡,马克龙
+6552421932208226819_!_113_!_news_world_!_通讯：政府资助 社会支持——挪威全民参与“海滩清洁日”活动_!_哺乳动物,海洋垃圾,海滩,海滩清洁日,挪威首都奥斯陆郊区
+6552379602319180035_!_113_!_news_world_!_日本和以色列为什么会在贫瘠的土地上建立起高度发达的国家？_!_
+6552327836424732935_!_113_!_news_world_!_怎么看因为性侵丑闻，二战后诺贝尔文学奖首次停摆？_!_
+6552387248652812547_!_113_!_news_world_!_俄罗斯军人很爱喝酒吗？_!_
+6552465029097062669_!_113_!_news_world_!_驱魔浪人在哪儿可以看？_!_
+6551245221026332941_!_113_!_news_world_!_以色列为什么最近要和伊朗针锋相对？_!_
+6552339585966276877_!_113_!_news_world_!_以色列宣称有秘密文件证明伊朗在核协议方面撒谎，你认为有多高可信度？_!_
+6552450140194472199_!_113_!_news_world_!_特朗普为什么可以随意废除此前美国政府达成的各种协议？_!_
+6552227896612618766_!_113_!_news_world_!_这次中美经贸磋商，美国终于认清这样一个现实_!_美方,倾巢出动,贸易战,大博弈,美国
+6552158260181336327_!_113_!_news_world_!_2018年美国中期选举，你认为特朗普会下台吗？_!_
+6552299178704240904_!_113_!_news_world_!_如果发生危机，要房子还是美元？_!_
+6550758697151234317_!_113_!_news_world_!_德国、英国和法国将派专家组前往以色列研究伊朗秘密核计划文件，普京该如何应对？_!_
+6552403283447644420_!_115_!_news_agriculture_!_在农村经常听人说“春栽杨柳夏栽桑，正月种松好时光”，这句话现在还适用吗？_!_
+6552462628692689412_!_115_!_news_agriculture_!_南阳：政企携手 创新模式 带动贫困户增收_!_牧原,联合社,南阳,内乡县,贫困户
+6552409752201069059_!_115_!_news_agriculture_!_总书记点赞的茌平耿店村，青年微信群里在聊啥_!_耿付征,大棚,住宅楼,茌平县,贾寨镇
+6552320736906707470_!_115_!_news_agriculture_!_2018年5月5日加州鲈、大黄鱼、牛蛙、太阳鱼、鲫鱼塘头价行情简析_!_浙江,太阳鱼,养殖户,大黄鱼,指环虫
+6552333767409664516_!_115_!_news_agriculture_!_美丽乡村——小丘镇移村_!_熏衣草,美丽乡村,铜川市,小丘镇,耀州区
+6552493598242242824_!_115_!_news_agriculture_!_假如回到农村生活，你会怎么养活自己？_!_
+6552456576114885127_!_115_!_news_agriculture_!_提个醒！8月1日起，没有这个证农资店将关门大吉_!_农药,农业部,农药经营许可证管理条例,农药经营许可证,经营许可证,农药管理,许可证
+6552253033315566094_!_115_!_news_agriculture_!_6旬庄稼汉凭一株红薯苗发家致富，一亩地挣到了10万元_!_魏丙午,林州市,张庆军,红薯,红薯苗
+6552230194386567693_!_115_!_news_agriculture_!_为什么父辈买养老保险，而子女不买就不可以领养老金？_!_社会保险,养老金,保险,养老,基本养老金
+6552374427655340552_!_115_!_news_agriculture_!_全国牛价又跌了！_!_利木赞牛,鲁西黄,公牛,新疆褐牛,西门塔尔
+6552483117393773069_!_115_!_news_agriculture_!_为什么越来越多的人喜欢去农家乐，因为他们生产的都是无公害蔬菜_!_东方ic,无公害蔬菜,农家乐,东方IC,温室大棚
+6552265667008004621_!_115_!_news_agriculture_!_“三宜”一体兴乡村 陕西岚皋大力实施扶贫搬迁_!_土坯房,岚皋县,堰门镇,刘兴春,芙蓉村
+6552460782116798980_!_115_!_news_agriculture_!_养老保险一次性补缴15年，60岁后每月可领1300元，你愿意交吗？_!_养老金,甘肃省,保险,基础养老金,农村
+6552406185507553544_!_115_!_news_agriculture_!_在家养龙虾，为什么几个月以后龙虾不见了？_!_
+6552433330724274702_!_115_!_news_agriculture_!_古老神秘的乌公侗寨竹米节，一家蒸竹米饭全寨香，你想尝尝吗？_!_侗族,竹米节,贵州省榕江县,东方IC,侗寨
+6552348277239447811_!_115_!_news_agriculture_!_农村的集市该不该取消呢？_!_
+6552235789667795470_!_115_!_news_agriculture_!_当好奔小康的“主心骨”_!_姚启忠,庐江县,党总支,水产养殖,柿树村
+6552300187321106947_!_115_!_news_agriculture_!_“废品”变“珍品” 村民进城看展品 后悔家里的老物件当废品丢弃_!_农村
+6551647451344797960_!_115_!_news_agriculture_!_怎么去非洲开农场，做生意？_!_
+6552390207155470852_!_116_!_news_game_!_LOL不离开高地也能获胜？疯狂玩法大战60分钟_!_超级兵,高地,ADC,lol,防御塔
+6552278051542008327_!_116_!_news_game_!_东方最新作《不可思议的幻想乡 莲花迷宫》公布 将登陆PS4/NS_!_幻想乡,Switch,Chunsoft,东方Project,PS4
+6552351062727066115_!_116_!_news_game_!_游戏世界：吃鸡风潮将过，这个网站又推出一款吃鸡页游引腾讯注意_!_7K7K小游戏,网页游戏,绝地求生刺激战场,绝地求生,极地求生,最终使命,绝地求生全军出击
+6552286957865533959_!_116_!_news_game_!_LPL四大ADC主观排名，UZI走下神坛但依旧是LPL第一AD_!_AD,LPL,EDG,ADC,UZI
+6552426816714309896_!_116_!_news_game_!_DNF中，你对你玩的职业有哪些不满的地方？_!_
+6552364547452174606_!_116_!_news_game_!_英雄联盟号令之旗在职业赛场上大火，难道就没有一个AP英雄能完美克制吗？_!_
+6552331781163450632_!_116_!_news_game_!_英雄联盟投降规则改动，15分钟4人同意就行，你怎么看？_!_
+6552416081581965837_!_116_!_news_game_!_王者荣耀：李白界的人气主播，一个李白打出各自风采_!_剑仙,王者荣耀,哈迪斯,李白主播,李逍遥
+6552250511213462030_!_116_!_news_game_!_LOL：三级时诺手被这三个英雄单杀，一点都不丢脸_!_Marin,武器大师,征服者,锐雯,诺手
+6552258837045313806_!_116_!_news_game_!_国产独立游戏失落城堡销量20万+，能带来什么启示？_!_
+6552273124014948868_!_116_!_news_game_!_《战神》海外直播热度高 观看时长接近2000万小时_!_Newzoo,战神,PS4,堡垒之夜,CS ： GO,守望先锋,Dota2,绝地求生,英雄联盟
+6552497567446336013_!_116_!_news_game_!_S11最强射手排名！第一名却只需2888金币？触手蓝烟用来打王者局_!_S11,S10,鲁班,射手,后羿
+6552348848251994631_!_116_!_news_game_!_刺激战场：光子现在都这么良心的吗？这空投让人两眼放光！_!_AWM,游戏,空投,刺激战场,吉利服
+6552390327783653892_!_116_!_news_game_!_那些经营不善而遗憾关服的手游之一 乐元素《梅露可物语》_!_乐元素,RPG游戏,梅露可物语,游戏的人,开心消消乐,天天爱消除,刀塔传奇,养成游戏
+6552276093569597959_!_116_!_news_game_!_魔兽世界杂谈：可怜的玛法里奥！你究竟还要凉多少版本？_!_玛法里奥,卡德加,德鲁伊,暗夜精灵,伊利丹
+6552355882531815694_!_116_!_news_game_!_王者荣耀怎么看赵云打野Alan的回归？_!_
+6552132548669997316_!_116_!_news_game_!_lol国内联赛、msi、洲际赛和s赛，这四项赛事哪个含金量最高？哪个最低？_!_
+6552446986686038276_!_116_!_news_game_!_《狂野飙车8》破解版在哪下？_!_
+6552392580271702279_!_116_!_news_game_!_《英雄联盟》选择大区后，游戏消失了，是怎么回事？_!_
+6552392995314860302_!_116_!_news_game_!_《英雄联盟》IG中单Rookie这名选手的水平到底怎么样？_!_
+6552448025166348557_!_101_!_news_culture_!_上联是“轻风细雨相思泪”，怎么对下联？_!_
+6552395007741919502_!_101_!_news_culture_!_上联:落笔驚风雨，怎么对下联？_!_
+6552299805840769294_!_101_!_news_culture_!_上联“一撇一捺写人生”，征下联？_!_
+6552329113103761678_!_101_!_news_culture_!_上联：一季花香游人醉，怎么对下联？_!_
+6552254838913106184_!_101_!_news_culture_!_上联：青松无衣不惧冷，如何对下联？_!_
+6552482387387744525_!_101_!_news_culture_!_马克思是世界上最伟大的哲学家吗？_!_
+6552312417198539021_!_101_!_news_culture_!_马克思是世界上最伟大的哲学家吗？_!_
+6552012958849827335_!_101_!_news_culture_!_丰子恺：艺术家对世间万物都要予以热诚的同情_!_丰子恺,漫画家,艺术家,画作
+6552107524772856324_!_101_!_news_culture_!_「中国制造」系列专题连载之——金漆木雕_!_木雕,金漆木雕,浅浮雕,樟树,金箔
+6552462347682709774_!_101_!_news_culture_!_上联：乳燕翻飞照碧水，如何对下联？_!_
+6548992837835292936_!_101_!_news_culture_!_我没钱只买得起盗版书，这可耻吗？_!_
+6551921575459815944_!_101_!_news_culture_!_建安七子之首王璨是在什么样的情况下创作了《登楼赋》？_!_建安七子,山东省,王粲,登楼赋,蔡邕
+6552111651489841672_!_101_!_news_culture_!_「四+六对联7」出句：笔尖纸薄，惯写人情世态；请您对句_!_人情世态,岁月风云,笔尖纸薄,惯写人情世态
+6552487885545668872_!_101_!_news_culture_!_为什么一定要做个善良的人？_!_
+6552348904849932808_!_101_!_news_culture_!_诗词欣赏：闺怨_!_王昌龄,闺怨
+6552425108714029319_!_101_!_news_culture_!_《说文解字》有什么证据说字的本义，不是臆想呢？_!_
+6552418916084220173_!_101_!_news_culture_!_上联：田在心上思故乡，求下联？_!_
+6552447358091657486_!_101_!_news_culture_!_上联: 残阳不语逐鸿雁，如何对下联？_!_
+6551700043500356110_!_101_!_news_culture_!_读《唐诗三百首》之七言绝句之李商隐_!_嫦娥,李商隐,贾谊,巴山夜雨,隋炀帝
+6552464803623862541_!_102_!_news_entertainment_!_倪萍和陈红，你喜欢谁？_!_
+6552454696571765262_!_102_!_news_entertainment_!_明星跌倒哪家强？范冰冰有人扶，奚梦瑶微笑，就服张韶涵的腾空飞_!_范冰冰,女明星,张韶涵,奚梦瑶,张韵涵
+6552291813946819079_!_102_!_news_entertainment_!_董子健王俊凯兄弟情：聊天表情包也要用小凯的！_!_萌萌哒,王俊凯,董子健,兄弟情
+6552383860259160580_!_102_!_news_entertainment_!_贾乃亮回归正剧 与毛卫宁合作《誓言》_!_产科男医生,贾乃亮,兵棋推演,毛卫宁,煮妇神探,刘奕君,发布会,誓言
+6552401296358375684_!_102_!_news_entertainment_!_如何评价《盲山》这样的电影？_!_
+6552345234620023304_!_102_!_news_entertainment_!_两位时尚辣妈，baby体重45.1，昆凌体重43.8，身材保持得真棒！_!_昆凌体重43.8,昆凌,时尚辣妈,angelababy,angelababy体重45.1
+6552478099575931140_!_102_!_news_entertainment_!_为什么人眼晴能一下就能看到几十亿光年的星星？_!_
+6552411567000912387_!_102_!_news_entertainment_!_《像我们一样年轻》杀青 陈翔陈瑶携手创业大餐_!_实力赛,姚晨,张晨光,陈翔,刘恩佑,月亮传记,如我们年轻
+6552342868160152078_!_102_!_news_entertainment_!_《演员的诞生》第二季确定暑期见，导师拟邀著名导演？_!_演员的诞生,凌潇肃,陈可辛,演员,浙江卫视
+6552337335285973518_!_102_!_news_entertainment_!_刘亦菲参加活动近照，不小心把咖啡泼导演身上，网友都替她心疼！_!_神仙姐姐,懂礼貌,陈嘉上,刘亦菲,娱乐圈
+6552285287676903950_!_102_!_news_entertainment_!_谢娜粉丝过亿获吉尼斯认证 总裁大呼她为“谢大伽”_!_吉尼斯世界纪录,菲律宾,快乐大本营,夏奇拉,谢娜,碧昂丝
+6551957542040240643_!_102_!_news_entertainment_!_为何刘嘉玲出现危机？昔日好友关之琳却这样回答！_!_娱乐圈,关之琳,花旦,刘嘉玲,粉丝
+6552272079754887683_!_102_!_news_entertainment_!_济南导演《胆笑心惊》在济首映，致敬《倩女幽魂》_!_首映礼,香港电影,济南,黄思婷,倩女幽魂,胆笑心惊,黄一飞
+6552244028878881287_!_102_!_news_entertainment_!_这是最好的漫威英雄，还是最差的漫威英雄？_!_黑豹,漫威,超级英雄,IMDB,奥巴马
+6552339039624626702_!_102_!_news_entertainment_!_迪丽热巴模仿刘亦菲穿起帅气的西装，你觉得谁会更好看呢？_!_热巴,西装,刘亦菲,迪丽热巴,迪丽
+6552270111376409095_!_102_!_news_entertainment_!_刘涛出席活动，现场被王自健揽腰抱起不撒手_!_刘涛,王自健
+6552286143323308301_!_102_!_news_entertainment_!_你如何看待《甄嬛传》中教导甄嬛侍奉皇帝的芳若姑姑，她是个什么样的人？_!_
+6552277103180513799_!_102_!_news_entertainment_!_她是女摇滚第一人，为恩师堕胎甚至想过自杀，如今不计前嫌成好友_!_斯琴格力乐,摒弃前嫌,臧天朔,恩师臧天朔,山歌好比春江水,斯琴格日乐
+1599795081350276_!_102_!_news_entertainment_!_我小时候喜欢阿拉蕾，因为带上眼镜长得像她，于是总被同学让我带眼镜模仿，结果久而久之，我就成了近视_!_阿拉蕾
+6552432837490901507_!_103_!_news_sports_!_世乒赛男团决赛：马龙樊振东许昕上场，奥恰洛夫受伤德国无奈变阵_!_排兵布阵,樊振东,奥恰洛夫,男团,乒乓球
+6552244229127536899_!_103_!_news_sports_!_如何看待“自由搏击与传统武术谁厉害”这样的问题？_!_
+6552326769393795341_!_103_!_news_sports_!_2018瑞典世乒赛男团半决赛，中国队3-0轻松取胜，如何评价三位球员的表现？_!_
+6552338377063006472_!_103_!_news_sports_!_2019年亚洲杯分组抽签揭晓，中国队首夺亚洲杯的机会真的来了吗？_!_
+6551631068443705860_!_103_!_news_sports_!_这一榜单乔丹也只能屈居第二 现役季后赛得分榜他已无可超越_!_热火队,得分,NBA,季后赛,FMVP
+6552128245213430286_!_103_!_news_sports_!_争冠关键战，老妇人能否主场擒敌_!_那不勒斯,意甲,尤文图斯,博洛尼亚,阿莱格里
+6552232102337708296_!_103_!_news_sports_!_女排北仑集训，被寄予厚望的小将李盈莹正在打磨哪些技术环节？_!_
+6552305051979743748_!_103_!_news_sports_!_乐福频繁一打三称霸猛龙内线 造就詹姆斯惊世绝杀_!_猛龙,詹姆斯
+6552294651842265352_!_103_!_news_sports_!_网传《马卡报》披露，内马尔希望加盟皇马搭档C罗，可能性有多大？_!_
+6552339001695535629_!_103_!_news_sports_!_主帅1言恐让德罗赞开启暴走？2数据扎心，绝境的他平平无奇！_!_猛龙队,德罗赞,勒布朗,骑士队,德韦恩
+6552299912535474445_!_103_!_news_sports_!_5.5号比赛恒大金英权解围踢到球了为什么裁判给点球？_!_
+6552473223953383940_!_103_!_news_sports_!_霸气！中国男队9-1击败德国 “王者之师”斩获世乒赛团体九连冠_!_樊振东,马龙,弗朗西斯卡,正手,许昕
+6552307387116552456_!_103_!_news_sports_!_骑士与猛龙G3，詹姆斯打板绝杀猛龙，这球是什么水平？_!_
+6552312022174794254_!_103_!_news_sports_!_广州富力外援扎哈维的孩子，球场秀球技，颇有老爸的风采_!_秀球技,球场秀球技,扎哈维,风采广州富力外援扎哈维,广州富力
+6552321800829665539_!_103_!_news_sports_!_刘诗雯在世乒赛女团决赛的表现如何？_!_
+6552243240240677379_!_103_!_news_sports_!_什么叫打球要用脑，看看人家隆指导_!_季后赛,鹈鹕队,隆多,米罗蒂奇,勇士
+6552294957493780749_!_103_!_news_sports_!_猛龙主教练跟泰伦卢在这个系列赛的执教水平你怎么看？_!_
+6552258428880814596_!_103_!_news_sports_!_身价仅值300万美刀却获中超1亿报价，瑞士巨星嘲讽这点钱太少了！_!_伊涅斯塔,力帆,伊布,中超,卡萨诺
+6552300457673359880_!_103_!_news_sports_!_天赋超越泰森，波兰猛将只因一大天生缺陷无缘世界拳王_!_戈洛塔,坏小子,拳击,泰森,刘易斯
+6552236214479487239_!_104_!_news_finance_!_现在股市都是2―3元的很多，有哪些可以长期持有？_!_
+6552392352592298254_!_104_!_news_finance_!_去香港买保险有何利弊？_!_
+6552264616561345038_!_104_!_news_finance_!_真相惊人！白马股闪崩频现 到底是赚钱机会还是割韭菜陷阱？_!_白马股,西水股份,中国平安,格力电器,海信电器,东阿阿胶,PEG,上证50,老板电器,伊利股份
+6552306670825898510_!_104_!_news_finance_!_美4月就业数据欠佳 国内经济短期承压_!_法国,欧元区,制造业PMI,美国,失业率
+6552354318874313224_!_104_!_news_finance_!_江苏开通首条国际邮轮航线_!_江苏连云港,邮轮航线,连云港,邮轮,辉煌号
+6552315924098056718_!_104_!_news_finance_!_安永：全球并购交易势头正旺不受风险影响 中国为首选五大投资目的地之一_!_地缘政治,机器人,并购,安永,高管,受访者
+6552305789342581252_!_104_!_news_finance_!_买了这只股票让巴菲特赚了七倍，而散户却至今还在深套之中_!_龙头股,中国石油H股,中国石油,巴菲特,伯克希尔哈撒韦公司,上证指数,伯克希尔哈撒韦
+6552391656417526285_!_104_!_news_finance_!_技术丨脱硫石膏在水泥生产使用中的几点措施_!_缓凝剂,脱硫石膏,水泥,水泥粉磨,辊压机
+6552452157654696452_!_104_!_news_finance_!_股东大会上，巴菲特还提到了减税、通货膨胀、报业转型_!_BNSF,阿里巴巴,巴菲特,伯克希尔·哈撒韦,母基金,碧桂园,股东大会,冰雪皇后
+6552486429589504519_!_104_!_news_finance_!_股神巴菲特也看多A股？_!_投资者,巴菲特,贸易战,白令海峡,股神巴菲特
+6552463143426064643_!_104_!_news_finance_!_50岁了，一直是电商方面自由职业者，现在准备职业炒股从哪里入手？_!_
+6552163767336043012_!_104_!_news_finance_!_招行呼和浩特分行感动服务添新举措_!_感动服务,金融,呼和浩特,营业网点
+6551958174931354125_!_104_!_news_finance_!_北京餐饮行业用工供、需居高 难堵用工缺口_!_北京餐饮行业,餐饮业,报告,餐饮企业,一线城市餐饮业,城市餐饮业
+6552404685888684292_!_104_!_news_finance_!_商标转让和商标变更一样吗？不一样吗？_!_
+6552367594484007437_!_104_!_news_finance_!_研究称2035年广东GDP将超英法 人均近20万元_!_广东GDP,广东省社会科学院,经济体,GDP,广东,研究
+6552241295186723336_!_104_!_news_finance_!_关于中国的投资机会、中美贸易 巴菲特在股东大会上是这样说的_!_特斯拉,中石油,亚马逊,Jain,伯克希尔哈撒韦,美国运通,伯克希尔·哈撒韦,中国石油,巴菲特,护城河,Abel
+6552241958943719940_!_104_!_news_finance_!_周一大盘机会难得不突破就创新低_!_大盘
+6552322814907515149_!_104_!_news_finance_!_如果一没有资金二没有技术，但是有创业的激情，应该如何创业？_!_
+6552135153357947399_!_104_!_news_finance_!_新闻简报：8.7亿！阿里首次公布支付宝们全球活跃用户数量_!_白洋淀,奢侈品,支付宝,乐视网,生态系统
+6552387438323433988_!_104_!_news_finance_!_中国股市常被股民骂，为什么？_!_自新之路,上证指数,股市,骂股市,中国股市
+6552398296311136525_!_106_!_news_house_!_中国楼市不断上涨，李嘉诚为什么加速撤离中国楼市？_!_
+6552305789292249614_!_106_!_news_house_!_“两微一抖”成标配？看房企如何玩转抖音_!_星座,成都龙湖,龙湖·天宸原著,抖音,星巴克
+6552364234720674055_!_106_!_news_house_!_房子漏水，物业不管，该怎么办？_!_
+6552451456039911687_!_106_!_news_house_!_婚房由男方支付首付和每月月供，如果离婚，女方能分到钱吗？_!_
+6552465459885637901_!_106_!_news_house_!_手上有25万，房价7800，可以首付买多大的房？_!_
+6552257986897641735_!_106_!_news_house_!_大连夫妇与中介公司口头确定买房后嫌中介费太高另投别家被打，你觉得中介辛苦介绍好房了之后变主意应该吗？_!_
+6552311852641026564_!_106_!_news_house_!_深圳房价迎19连跌！上月跌6元 这月再跌16元_!_均价,三价合一,深圳,各区,租赁
+6552279466297524749_!_106_!_news_house_!_芜湖购房最高可补贴200万元 再出人才新政_!_战略性新兴产业,芜湖市,芜湖,购房,人才引进
+6552477326687339012_!_106_!_news_house_!_贷款买房前 你是否知道这几大误区了呢？_!_购房者,提前还贷,住房公积金,借款人,公积金贷款
+6552246662415253767_!_106_!_news_house_!_“现在抓紧买房，十年以后买不起”，对于这种说法你怎么看？_!_
+6552286701228655107_!_106_!_news_house_!_一季度广州商品房销售面积同比现“腰斩”_!_珠三角,房地产开发企业,商品房销售面积,广东商品房销售面积,办公楼,广东
+6552331464837431559_!_106_!_news_house_!_2018年成都房价还会涨吗？_!_
+6552398567334478343_!_106_!_news_house_!_嘉兴秀洲中安银领国际，房价低的没底线，能买吗？_!_秀洲中安银领国际房价,嘉兴秀洲中安银领国际房价,LOFT,中安银领国际房价,嘉兴秀洲中安银领国际
+6552295319999086855_!_106_!_news_house_!_海南琼中买的房子，限购政策只能卖给本县人，以后还能卖出去吗？_!_
+6552419279365472526_!_106_!_news_house_!_错过了楼市，错过了比特币，下一个风口会是什么？_!_
+6552395148754420237_!_106_!_news_house_!_今天，马云入局，房地产行业将迎来毁灭性洗牌！_!_Ma Yun,房地产市场,二手房,租赁市场,Ali
+6552422535965704462_!_106_!_news_house_!_手中资金大约600万闲置资金，房产已限购，该怎么配置投资？_!_
+6552433857893761287_!_106_!_news_house_!_海南房地产为什么会限购？_!_
+6552235499027694088_!_107_!_news_car_!_本田思域将推出SI版，外观更偏跑车设计，碳纤维中控，依然搭1.5T_!_本田思域,思域,思域SI,CR-V,跑车
+6552393221605949703_!_107_!_news_car_!_预算15万，个子矮的女生适合开什么车？_!_
+6552372044409864708_!_107_!_news_car_!_宝马车被贴满女性用品 监控拍下深夜恶作剧女子 究竟有何深仇大恨_!_宝马,用品,宝马车,恶作剧,女性
+6552150453654651395_!_107_!_news_car_!_李书福这次火了！国产“路虎揽胜”来袭，不足20万，买途观后悔_!_路虎揽胜,李书福,揽胜,吉利,汽车品牌,外观设计,路虎,Concept
+6552486768371826952_!_107_!_news_car_!_途观L和x1，哪个好？_!_
+6552418197675442435_!_107_!_news_car_!_你开过最快的车速是多少？_!_
+6552398029813448973_!_107_!_news_car_!_汽车一定要在4S店贴膜吗？_!_
+6552145799663321613_!_107_!_news_car_!_车上零碎物品多没地放，车载座椅夹缝置物盒，不占空间还方便_!_密封条,座椅,置物盒,迫在眉睫,收纳盒
+6552339503380431108_!_107_!_news_car_!_荣威Tx3、瑞虎7、帝豪gs1.8T和吉利帝豪gl哪个好？为什么？_!_
+6552308325113922052_!_107_!_news_car_!_花5万买辆丰田普拉多，这车搭载V6发动机，所以觉得一点都不亏_!_普拉多,SUV,丰田普拉多
+6552367713124090381_!_107_!_news_car_!_本想买卡罗拉的我，最后却提了这部suv，只因这车性价比太高了！_!_消音器,ix35,X80,吉利博越,性价比,国产车,卡罗拉,xr-v
+6552339056221487619_!_107_!_news_car_!_荣威RX8试驾，一顿操作猛如虎！_!_RX3,上汽荣威,RX5,RX8,GL8,哈弗H9,CS95,GS8,荣威,荣威RX8,SUV
+6552293444524442120_!_107_!_news_car_!_想买速腾的再等等！这车颜值逆天，溜背造型，比帕萨特便宜好几万_!_帕萨特,朗逸,速腾,Carplay,朗逸Plus,高配车型
+6552369616394387982_!_107_!_news_car_!_居全国GDP最高的省份，风景优美却物价低廉，节假日酒店从不加价_!_大酒店,张资平,客天下,梅州十八景,留余堂
+6552399644385607940_!_107_!_news_car_!_轮胎需要定期调换吗？_!_
+6552393518457815556_!_107_!_news_car_!_一汽拼了！新车比奥迪漂亮十倍，造型个性设计前卫，仅卖8万！_!_奔腾T77概念车,一汽奔腾T77概念车,奔腾T77,一汽奔腾,一汽奔腾T77,一汽,奔腾
+6552378825731211783_!_107_!_news_car_!_还能这样操作？停车被贴罚单！车主：我是停在自己家门口！_!_道路交通平安法,机动车,人行道,交警叔叔,家门口
+6552345164864553229_!_107_!_news_car_!_奔驰s350显示动力转向故障，挂不起挡位，是怎么回事？_!_
+6552321961647669767_!_107_!_news_car_!_日内瓦国际汽车展览会_!_汽车展览,超级跑车,International,Auto,日内瓦
+6552315144188199428_!_107_!_news_car_!_买了七座车之后，很多人后悔了！_!_MPV,SUV,第三排,车管所,汉兰达,座车
+6552251396555538695_!_108_!_news_edu_!_如果你是一名教师，同意取消职称吗？_!_
+6552359261127573764_!_108_!_news_edu_!_当年你的那些“死读书”的同学，现在都怎么样了？_!_
+6552362186348429837_!_108_!_news_edu_!_自考学历社会认可度如何？_!_IBM,吴士宏,高等教育自学考试暂行条例,自考学历,张静初,含金量
+6552345469375217924_!_108_!_news_edu_!_2018国考跟省考同时入面试怎么办呢？_!_
+6552285468568846862_!_108_!_news_edu_!_高考试卷中的极化恒等式_!_极化恒等式,数量积,恒等式
+6552245596302868999_!_108_!_news_edu_!_高考“白卷”考生10年后考取大专 后悔浪费高考机会_!_高考,幡然悔悟,教育宣言,专升本,徐孟南
+6552300871554695684_!_108_!_news_edu_!_Rich Brian expected at the Westlake Music Festival!_!_Brian,February,Westlake,Indonesian,Rich
+6552273939492503821_!_108_!_news_edu_!_大学中庸小五月，如何接句？_!_
+6552462560451363085_!_108_!_news_edu_!_美国是怎样进行爱国主义教育的？_!_
+6552350001689788679_!_108_!_news_edu_!_教育的分类有哪些？_!_
+6552496493998113031_!_108_!_news_edu_!_哪些大学的研究生院工程管理专业还可以？_!_
+6552418430010524174_!_108_!_news_edu_!_你会念这四个单词么？鸿鹄之志，莘莘学子，乳臭未干，谆谆教诲_!_
+6552393150269227534_!_108_!_news_edu_!_绩效比教师高2000多，谁拿走了老师绩效？_!_一线教师,教导处,学期绩效工资,中心校,总务处
+6552412470802120968_!_108_!_news_edu_!_地处合肥的中国科学技术大学与地处武汉的华中科技大学相比而言，哪个略胜一筹？_!_
+6552276474706002436_!_108_!_news_edu_!_手指是大脑发育的关键，这样锻炼，娃想不聪明都难_!_大脑发育,蒙台梭利,岁宝宝,个月宝宝,月宝宝手部发育
+6552293323481022990_!_108_!_news_edu_!_高考倒计时！渤海中学高三学生“立下军令状”为高考“冲刺”_!_滨州渤海中学,渤海中学,东北镇,焦点访谈,中央电视台
+6552396895287771651_!_108_!_news_edu_!_「会计初级资格证备考」天天上班加班将如何备考呢？_!_资格考试,学习方法,资格证,会计专业,会计专业技术初级
+6552491631273050382_!_108_!_news_edu_!_重庆所有高中排名，你觉得前五是哪几所？_!_
+6552311352260559117_!_108_!_news_edu_!_音乐学学渣考研有多大的希望？_!_
+6552253941059420423_!_109_!_news_tech_!_一加6手机需要抢吗？一加的系统和小米比怎么样？_!_
+6552471524241047811_!_109_!_news_tech_!_怎么看待有人冒充马化腾回应《腾讯没有梦想》一文？_!_
+6552312384453607688_!_109_!_news_tech_!_小白如何提高苹果手机拍摄的技巧？_!_
+6552461743862317320_!_109_!_news_tech_!_智慧农业大棚监控系统能实现哪些功能？_!_
+6552361867417747971_!_109_!_news_tech_!_小米雷军和格力董明珠赌约到期：全国人民作证的十亿赌注算数吗？_!_小米,董明珠,雷军,香港,小米公司
+6552345840076194062_!_109_!_news_tech_!_现在很多人做淘宝优惠券，如何盈利的？_!_
+6552367589878661636_!_109_!_news_tech_!_营收是阿里的两倍，市值等于三个沃尔玛，利润却只有阿里三分之一_!_贝佐斯,云计算,阿里,亚马逊,沃尔玛
+6552430651914584334_!_109_!_news_tech_!_马化腾为什么不把南非的股份收回来？_!_
+6552353504034292228_!_109_!_news_tech_!_一夜之间，马云放大招，芝麻分700，网商贷可借百万_!_网商,网商贷,马云,借呗,支付宝
+6552263448619647501_!_109_!_news_tech_!_美国人抱怨：“为什么美国没有普及手机支付？瞧瞧人家中国！”_!_信用卡,支付宝,银行业,互联网,美国
+6552469782250127623_!_109_!_news_tech_!_10万创业？_!_
+6552302793045049859_!_109_!_news_tech_!_马化腾都站台小程序，你觉得没机会，腾讯连接一切的野心就摆那里_!_微信,马化腾,程序,公众号,OPPO
+6552399723406295300_!_109_!_news_tech_!_成功人士用华为，穷人用苹果，屌丝用小米，事实果真如此吗？_!_
+6552330028112150792_!_109_!_news_tech_!_为什么中国不能像美国针对华为手机一样也针对美国苹果手机禁售？_!_
+6552400913233871107_!_109_!_news_tech_!_潜伏如果陆桥山不死，会帮助余则成当上站长吗？_!_
+6552298295731962125_!_109_!_news_tech_!_小米要上市了，有十万块钱怎么能买到小米的股票？_!_
+6552277293765493252_!_109_!_news_tech_!_马化腾回应《腾讯没有梦想》：我的理想不是赚多少钱，是做出最好的产品_!_腾讯没有梦想,核心优势,马化腾,事业线,腾讯
+6552381721512247565_!_109_!_news_tech_!_腾讯市值超过阿里，但为什么在各种场合上，总是感觉阿里格局更高？_!_
+6552238943855706375_!_109_!_news_tech_!_中国人一年不买苹果手机，苹果会破产吗？_!_
+6552078597861933572_!_110_!_news_military_!_央视曝光：中国研制成功下代超隐身材料！美国：将领先美国一代！_!_歼11,歼-15,歼-11,超材料,F-22,F35,F22,歼15,F-35,隐身材料
+6552225528676352269_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552260277461582084_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552095189903606279_!_110_!_news_military_!_五角大楼神秘办公室面临废黜 如今或让“第二神盾局”接管_!_DIUx,美国国防部,美军,国防部长,五角大楼
+6552447159071932685_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552465098202415367_!_110_!_news_military_!_波兰国防部长为何称美国驻军是波兰梦想了几十年才完成的事情？_!_
+6552272978971721987_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552237045907980547_!_110_!_news_military_!_中国的辽宁航母战斗群的战力如何？_!_
+6537435759190737165_!_110_!_news_military_!_一发155榴弹击中M1A1坦克的正面装甲能否将其摧毁？_!_
+6552324227083534606_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552315474086985991_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552399184140435716_!_110_!_news_military_!_你怎么看待以色列总理出席俄罗斯5月9日的阅兵并会晤普京？_!_
+6552131365947572487_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552496290985410819_!_110_!_news_military_!_如果美国退出联合国怎么办？_!_
+6552471325464592654_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6551547636007043341_!_110_!_news_military_!_德国为什么会在第一次世界大战中战败？_!_
+6552400482868920840_!_110_!_news_military_!_无奋斗不青春｜大火箭总装团队：我为祖国铸利箭_!_无奋斗不青春,运载火箭,火箭,长征五号,总装,整流罩
+6552412971195187470_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6550432036228694285_!_110_!_news_military_!_二战中的不列颠空战对德国来说是否具有战略意义？_!_
+6552405260780962312_!_112_!_news_travel_!_西藏一季度接待游客人次和总收入增长均超四成_!_林芝,日喀则,冬游西藏,巴松措,西藏自治区
+6552279346764055047_!_112_!_news_travel_!_越南语学习篇之越南语中常用词语三十二，建筑行业词语翻译表格_!_越南语,旅游攻略,建筑行业
+6552424837027987976_!_112_!_news_travel_!_福利几款旅行必备神器，便宜实用颜值高！_!_便携式,洗漱包,颜值,充气枕
+6552440972846825992_!_112_!_news_travel_!_幽鹤公园，加拿大国家公园中以壮观著称的一个，面积不大风景极好_!_天生桥,路易斯湖,印第安语,寒武纪,幽鹤
+6552342774467789316_!_112_!_news_travel_!_贵州毕节的这几处溶洞，夏天探险的好去处_!_织金洞,九洞天,洞口,卷曲石,千姿百态
+6552358570908713223_!_112_!_news_travel_!_四川游客张某带着六名女子在三亚沙滩烧烤被撒沙，这事你怎么看？_!_
+6552279868732604941_!_112_!_news_travel_!_唐汪镇全力提升旅游产业_!_洮河,国庆节,杏子,唐汪川,唐汪镇
+6552190754154873102_!_112_!_news_travel_!_日落霞飞苍山远，求下句？_!_
+6552150207528698372_!_112_!_news_travel_!_初见倾心，我在东部华侨城度过了美好的一天_!_大峡谷,东部华侨城,观光塔,高尔夫球场,同心锁
+6552211505733960206_!_112_!_news_travel_!_带雨伞太麻烦，越来越多女人出门包里放个胶囊片，刮风下雨都不怕_!_榨汁机,旅行睡袋,电动车,五折伞,三合一
+6552295633762386440_!_112_!_news_travel_!_美媒报道中国40个最美丽最值得游览的景点，这里有你喜欢的吗？_!_武夷山,黄山,云冈石窟,月牙湖,承德避暑山庄
+6552329690898497796_!_112_!_news_travel_!_内蒙古呼伦贝尔大草原好玩吗？几月份去好呢？_!_
+6552201516785074440_!_112_!_news_travel_!_什么样的炭燃烧时间最长？_!_
+6552485918027022596_!_112_!_news_travel_!_四川游客张某带着六名女子在三亚沙滩烧烤被撒沙，这事你怎么看？_!_
+6552280426860249607_!_112_!_news_travel_!_欧洲庄园有什么好羡慕的？种些花草就可以攀比了_!_蝶形花冠
+6552436045244268808_!_112_!_news_travel_!_西安的特产在哪可以买到？_!_
+6552388351008178692_!_112_!_news_travel_!_海南法国维希小镇：23万一套起 低密度果岭度假社区 享受静懿生活_!_法国,享受静懿生活海南法国,果岭度假社区享受静懿,海南法国维希小镇,密度果岭度假社区享受
+6552420610398486788_!_112_!_news_travel_!_世界第一伟人是谁？_!_
+6552394512411394573_!_112_!_news_travel_!_陕西这几个美得惊艳的地方，雄伟奇险 但是去的人很少！_!_镇安,淳化县,秦岭,宁陕县,塔云山
+6552365975885316612_!_112_!_news_travel_!_雨花微旅行 圭塘河岸_!_普罗旺斯,微旅行,圭塘河岸,浏阳河,雨花区
+6552319671041786125_!_113_!_news_world_!_美国法院判伊朗赔付9·11遇难者家属逾60亿美元，你怎么看？_!_
+6552301359440331015_!_113_!_news_world_!_俄罗斯为什么斗不过美国？_!_
+6552123582820909326_!_113_!_news_world_!_作为俄罗斯总统，普京上台以来有哪些失误？_!_
+6552486899334775053_!_113_!_news_world_!_非洲为什么没有熊？_!_
+6552404054981476611_!_113_!_news_world_!_如果证实反政府武装使用化学武器，美英法会被追责吗？_!_
+6552307196921643534_!_113_!_news_world_!_特朗普助选重申“美国第一”，注定只能帮倒忙_!_特朗普,孤立主义,共和党,美国第一,俄亥俄州
+6552037534711415047_!_113_!_news_world_!_为什么英国组建“反俄联盟”？_!_
+6552354429016736007_!_113_!_news_world_!_为何奥巴马刚上台就获诺贝尔和平奖？他给世界带来和平了吗？_!_
+6551277975759749645_!_113_!_news_world_!_俄罗斯警察的腐败程度有多高，说出来可能都没人相信_!_俄罗斯警察,俄罗斯,小费,警察
+6552413557168800269_!_113_!_news_world_!_中国女人嫁非洲男人都嗤之以鼻，那中国爷们娶非洲女人你们怎么说_!_两相情愿,中国妹子,嫁给黑人,走出非洲,棕色人种
+6552446435021816334_!_113_!_news_world_!_美患癌资深参议员术后出院 提前安排后事葬礼不欢迎特朗普_!_特朗普,共和党,麦凯恩,亚利桑那州,奥巴马
+6550919810807496963_!_113_!_news_world_!_为什么日本人很少有艾滋感染者？_!_
+6552353546921050632_!_113_!_news_world_!_特朗普定下与金正恩会面时间地点 并邀文在寅访美_!_得克萨斯州,文在寅,金正恩,青瓦台,驻韩美军
+6551291905492123918_!_113_!_news_world_!_为什么一个国家不能无限印钞票来让自己有钱，购买更多国外资源进行本国建设？_!_
+6552380368962453768_!_113_!_news_world_!_韩国本国公民去济洲岛需要持什么证件？_!_
+6552271901043982856_!_113_!_news_world_!_美国又栽个跟头！叙利亚化武袭击新进展：这个秘密组织成替罪羊！_!_美国,化武袭击,俄罗斯,化武,袭击事件
+6551943119267430925_!_113_!_news_world_!_叙利亚还要乱多久？俄罗斯：前景堪忧，美国实乃罪魁祸首_!_反对派,美国,阿拉伯之春,叙利亚,巴沙尔
+6552286521561448967_!_113_!_news_world_!_加沙发生爆炸致死6人_!_哈马斯,加沙地带,阿什拉夫,卡德拉,卡桑旅
+6552248283736047880_!_113_!_news_world_!_如果证实反政府武装使用化学武器，美英法会被追责吗？_!_
+6552155850801152264_!_115_!_news_agriculture_!_本人开火锅店，想加入麻辣小龙虾，哪里有小龙虾养殖基地？_!_
+6552425703386645000_!_115_!_news_agriculture_!_刚刚！达州宣布成功产出环保新材料：比头发丝还细！每吨卖2万！_!_达州,玄武岩纤维原丝,玄武岩纤维,玄武岩
+6552409237870346766_!_115_!_news_agriculture_!_告诉你为什么城里买不到真正的农村土鸡蛋_!_土鸡,农村,土鸡蛋,人工费,母鸡
+6552455811400991240_!_115_!_news_agriculture_!_过完年从农村老家出来，一直没回家，想孩子了_!_农村
+6552383898658013703_!_115_!_news_agriculture_!_村里“最为难”的3种人，要提前避免这种遭遇_!_农村
+6552362994670502414_!_115_!_news_agriculture_!_今年大豆要扩种，补贴大幅度提高哦！_!_大豆,进口量,大豆种植,大豆生产者补贴,大豆种植面积
+6552398376602698243_!_115_!_news_agriculture_!_趣谈文化：你意想不到的奇怪习俗，韩国用它来当作大礼_!_身土不二,肉类食品,韩国,牛肉,农业
+6552275387957641735_!_115_!_news_agriculture_!_立夏后的农事概要，农民朋友们必须知道_!_立夏,百分之,花芽分化,病虫害,红蜘蛛
+6552466174959944205_!_115_!_news_agriculture_!_农民注意！2018年城市里的农村人要想“回农村”，可参考以下标准_!_农转非,农村,非转农,农村户口,农业户口
+6552287774311973380_!_115_!_news_agriculture_!_乡村振兴需要工匠精神_!_工匠精神,农村,民间匠人领路乡村振兴潮,及时雨,匠人,乡村振兴
+6552449637200953860_!_115_!_news_agriculture_!_这种长得像猫爪子的野菜，市场空间大，成了农民致富的“摇钱树”_!_唐松草,长白山,猫爪子菜,野菜,摇钱树
+6552320045207257607_!_115_!_news_agriculture_!_李晖：切实帮助搬迁贫困户融入新环境 致富奔小康_!_搬迁贫困户,易地扶贫搬迁,九疑山,宁远县,李晖
+6552328488160854531_!_115_!_news_agriculture_!_如何判定一头公猪应该淘汰了？应当给予其四次证明自己的机会_!_种公猪,母猪
+6552465324459950605_!_115_!_news_agriculture_!_湖北宜昌：大雨之后，天空酷似水墨画！乡村农房相映美！_!_天空酷似水墨画,宜昌市,仓屋榜村,水墨画,湖北
+6552449741270024708_!_115_!_news_agriculture_!_水产疾病频发、养殖成功率降低、产量急剧下降，水质恶化怎么办？_!_微生态制剂,水产养殖,养殖,悬浊液,絮凝作用
+6552410272114409731_!_115_!_news_agriculture_!_水泥钢筋价格大幅度的上涨，农村自建房怎么办？_!_
+6552419291507982851_!_115_!_news_agriculture_!_福建农村电网改造助力乡村振兴_!_变压器,供电公司,福建,千伏安,电网
+6552308142187741709_!_115_!_news_agriculture_!_聪明女人都不在家养花了，改种这几种果苗，年年水果吃不完_!_火龙果,雌雄异株,西番莲,树苗,百香果
+6552279590964822542_!_115_!_news_agriculture_!_崇义筑牢健康扶贫“四道保障线”_!_疾病,崇义县,道保障线,贫困人口,疾病医疗补充
+6552427200711229959_!_115_!_news_agriculture_!_这些水果算是比较罕见的，一般地方很难遇到，吃过四种就不得了_!_土耳其,山竹,水果,巧克力,莽吉柿
+6552474077292921347_!_101_!_news_culture_!_康熙祭拜孔子没有跪下，大臣挡住墓碑上一个字，康熙立即下跪叩拜_!_葛尔丹,康熙,孔子墓,孔子,孔子庙
+6552268820306723079_!_101_!_news_culture_!_中国十八世纪时，都有哪些著名的诗人？_!_
+6552375588579967245_!_101_!_news_culture_!_上联：仙山升明月，下联怎么对？_!_
+6552269384977481997_!_101_!_news_culture_!_“花笑夕阳暮”下一句怎么接？_!_
+6552257854571545091_!_101_!_news_culture_!_渭南新增少华山传说等7个省级非遗保护项目_!_少华山,武帝庙,陕西省,非物质文化遗产,麦秆画
+6552392068256236046_!_101_!_news_culture_!_留住乡愁！湛江最美古村落要消失了，抓紧机会看一回吧_!_古建筑,古民居,雷州,荔枝村,苏东坡
+6552386369178566925_!_101_!_news_culture_!_看完哪些书会让你觉得作者的知识量真的很大？_!_
+6552384829390848259_!_101_!_news_culture_!_为什么当代青年评选“十大读不下去的书”，《红楼梦》排名榜首？_!_
+6552471518381621768_!_101_!_news_culture_!_千古爱情故事系列之：不思量自难忘——东坡与王弗的爱情故事_!_唤鱼池,王弗,王朝云,苏轼,王方
+6552336834603516164_!_101_!_news_culture_!_马克思是世界上最伟大的哲学家吗？_!_
+6552331404082938120_!_101_!_news_culture_!_上联：柳乱池塘水，下联怎么对？_!_
+6552181460726448392_!_101_!_news_culture_!_上联:十面埋伏刀斧手，下联？_!_
+6552378835315196429_!_101_!_news_culture_!_春天的北京箭扣长城，冰雪奇缘_!_望京楼,箭扣长城,山海关,慕田峪长城,九眼楼
+6552494047666110728_!_101_!_news_culture_!_上联：解手，下联如何对？_!_
+6552255601710203144_!_101_!_news_culture_!_下联：游古镇，抚古筝，古意正浓。怎样对上联？_!_
+6552275215886319876_!_101_!_news_culture_!_太古，请赐联？_!_
+6552172203939463693_!_101_!_news_culture_!_中国赠送马克思雕像在德国特里尔揭幕_!_特里尔,吴为山,雕像,马克思,史明德
+6552312626834047236_!_101_!_news_culture_!_有1700年历史记载的隆中和有700年历史记载的卧龙岗，哪个更可能是诸葛亮的躬耕地？_!_
+6552377711870870024_!_102_!_news_entertainment_!_迪丽热巴的脚，杨幂的脚，刘亦菲的脚，在她面前都自惭形秽_!_上海戏剧学院表演系,迪丽热巴,影视女演员,流行乐,上海师范大学
+6552219248633053447_!_102_!_news_entertainment_!_明星看自己演的电影电视剧是什么感觉呢？_!_
+6552488518554223112_!_102_!_news_entertainment_!_中国最难请的4位明星，第2无人不知，第4无人能请动！_!_中华武术,元秋,制作人,周星驰,星辉公司,战狼2,功夫
+6552273394732105998_!_102_!_news_entertainment_!_华晨宇除音乐之外还有哪些可塑性？_!_
+6552417663186895363_!_102_!_news_entertainment_!_15年过去了，《倚天屠龙记》演员颜值对比，有的似大妈有的如少女_!_妈妈是超人,陶虹,苏有朋,贾静雯,高圆圆,陈秀丽
+6552262750985257479_!_102_!_news_entertainment_!_Angelababy出席活动，网友；模特出道的你，演戏一样不比别人差_!_Angelababy,angelababy,模特,模特出道,Angelababy出席活动
+6552371168257507854_!_102_!_news_entertainment_!_陈冰出席发布会，网友：就服你这二郎腿，实在是太有个性_!_陈冰,个性陈冰出席发布会
+6552446202430882317_!_102_!_news_entertainment_!_蔡徐坤、范丞丞公关做得太浮夸，这一波微博数据一看就很假！_!_范丞丞,微博,转发量,蔡徐坤,谢娜,偶像练习生
+6552421265217421575_!_102_!_news_entertainment_!_如何评价袁立？_!_
+6552492365280444942_!_102_!_news_entertainment_!_鹿晗前女友曝光，比关晓彤还要性感漂亮 王思聪：你应该感谢我_!_奔跑吧兄弟,关晓彤,迪丽热巴,王思聪,鹿晗,撒狗粮
+6552364772254286343_!_102_!_news_entertainment_!_宋朝的家国情怀_!_天将雄狮,华莱士,出塞曲,角斗士,电影,琅琊榜,辛德勒,黄金甲,阿凡达,大风歌,金陵十三钗,价值观,造型师,勇敢的心,马特达蒙,奥斯卡,长城
+6552343091787858435_!_102_!_news_entertainment_!_明星也爱吃棒棒糖，赵丽颖好甜，吴亦凡霸气，杨洋太魔性了_!_棒棒糖,吃棒棒糖,吴亦凡,刘亦菲,赵丽颖
+6552270266611794180_!_102_!_news_entertainment_!_为什么在《复仇者联盟3》中地球的军方没有参战？_!_
+6552135140703731975_!_102_!_news_entertainment_!_《温暖的弦》原本是由黄晓明担任该剧男主角，可现在却被换成了张翰，这事你怎么看？_!_
+6552059437891191300_!_102_!_news_entertainment_!_何炅为谢娜冒雨进村买鸡蛋？网友知道原因后被暖哭：这是一筐友情_!_谢娜,何炅,快乐大本营
+6552428525196935694_!_102_!_news_entertainment_!_陈小春一家三口重游结婚旧地，Jasper玩的非常开心，太可爱了！_!_拉斯维加斯,爸爸去哪儿,陈小春,真人秀,Jasper
+6552338542763180552_!_102_!_news_entertainment_!_马克思思想之光照进银幕荧屏_!_恩格斯,资本论,王一川,纪录片,青年马克思,马克思主义,马克思,共产党宣言
+6552307872380748302_!_103_!_news_sports_!_大荔县朝邑镇：乘“大马”东风 展名镇风采_!_马拉松,大马,马拉松赛,朝邑镇,丰图义仓
+6552310124730384904_!_103_!_news_sports_!_詹姆斯又双叒叕绝杀，洛瑞直接懵逼_!_乔丹,洛瑞,猛龙,三分线,詹姆斯
+6552245129258729736_!_103_!_news_sports_!_瑞典世乒赛男团决赛，中国队面对德国队，应该派出怎么样的阵容？_!_
+6552303999578538510_!_103_!_news_sports_!_真急了！詹姆斯与洛瑞发生冲突，三分回应再现死亡之瞪+绝杀比赛_!_猛龙,洛瑞,NBA,詹姆斯,伊巴卡
+6552130612486996493_!_103_!_news_sports_!_皆大欢喜！国乒女队夺冠后接受采访 中肯评价日本女乒发挥_!_朱雨玲,丁宁,刘诗雯,平野美宇,伊藤美诚
+6552285551246967299_!_103_!_news_sports_!_朱婷受伤下场调整，瓦基弗银行差点被逆转，朱婷决胜局力挽狂澜_!_瓦基弗银行,朱婷,局瓦基弗银行,科内利亚诺,罗马尼亚
+6552409065023078926_!_103_!_news_sports_!_国家德比来袭！巴萨力争胜利送别大师，皇马却摆出全替补阵容？_!_伊涅斯塔,皇马,国家德比,巴萨,欧冠
+6552457927817429517_!_103_!_news_sports_!_中甲最新积分榜：梅县铁汉21分钟连入6球打爆辽宁宏运升至第七_!_梅县,辽宁宏运,阿洛伊西奥,梅县铁汉,穆里奇
+6552310995178488078_!_103_!_news_sports_!_奥尼尔:艾科麦佛鲨VS维韦杜詹霍，谁是更强战队？_!_
+6552425411479863560_!_103_!_news_sports_!_2019亚洲杯中国国足到底怎样做才能“夺冠”？_!_
+6552481364246004228_!_103_!_news_sports_!_埃梅里：姆巴佩本来想去皇马或巴萨，但我劝住了他_!_法国人,姆巴佩,埃梅里,皇马,巴萨
+6552321168223764750_!_103_!_news_sports_!_亚洲杯分组国足抽得上上签，国足就一定能出线吗？_!_
+6552353001619587592_!_103_!_news_sports_!_拒绝握手海因克斯 莱万多夫斯基究竟为何？_!_曼朱基奇,莱万,拜仁,莱万多夫斯基,海因克斯
+6552278241778860547_!_103_!_news_sports_!_NBA最新票价排名！勇士队第一，火箭队看了都尴尬_!_常规赛,老鹰队,勇士队,火箭队,美金排名
+6552326796988121347_!_103_!_news_sports_!_NBA最弱球队和CBA最强球队打比赛，哪个能赢？_!_
+6552425055588975118_!_103_!_news_sports_!_德罗巴解密助攻秘诀：我喝了球童的圣水_!_德罗巴,圣水,洛杉矶,赛后,约翰逊
+6552854889205072388_!_103_!_news_sports_!_国足携手杰尼亚！里皮推选3小将前往罗马训练_!_中国之队,杰尼亚,中国足球协会,中国足球,Zegna
+6552406891144675597_!_103_!_news_sports_!_如何在商业活动中不受人欺骗？_!_
+6552401223599784456_!_104_!_news_finance_!_全自动炒币机器人CCR炒币软件原理介绍_!_机器人,交易所
+6552478442405757454_!_104_!_news_finance_!_三百温商共同发起，为温州企业家量身打造 “慧商大学”正式挂牌_!_慧商大学,总裁慧客厅,蒋文广,企业家,共享经济
+6552371533606552067_!_104_!_news_finance_!_额度这么低，信用卡怎么提到10万呢_!_中国银行,信用卡,临时额度,银行,提额,浦发银行,农业银行
+6552274290526388494_!_104_!_news_finance_!_股票增发和股票回购是正好相反的两种操作吗？_!_
+6552249557344846350_!_104_!_news_finance_!_05月06日潜江天气预报_!_最低气温,潜江,中国天气网,摄氏度,东北风
+6552362889603187214_!_104_!_news_finance_!_李谨詺：周一该如何布局，5.7黄金不破1318短期震荡下行为主_!_谨詺,操作,黄金,单子,李谨詺
+6552285505814266125_!_104_!_news_finance_!_都说80年下海、90年炒股、00年炒房、10年炒比特币，那么现在该干些什么？_!_
+6552273609820209412_!_104_!_news_finance_!_建筑工程中同一项目，母公司投标，子公司能否投标？_!_
+6552488596719272206_!_104_!_news_finance_!_大智慧烧14亿给李大霄罗杰斯做直播，结果亏17亿要退市了，这是谁的锅？_!_
+6552415170486862349_!_104_!_news_finance_!_碧生源(00926)预计2018年上半年收入大幅下降 预亏2000万-3000万元_!_碧生源,分销商
+6552303431892074760_!_104_!_news_finance_!_合阳县未来有什么发展？_!_
+6552419877125095949_!_104_!_news_finance_!_方法论｜大小盘基金组合再探：中美两国股市的组合变通比较_!_沪深300,罗素2000指数,小盘股,中证800,中证,大盘股,亚马逊,中证500,罗素,指数基金
+6552325345255620871_!_104_!_news_finance_!_芝麻信用是怎么计算出来的？_!_
+6552137622179807757_!_104_!_news_finance_!_最新爆文｜存银行20万一年后亏了500，必须看看是怎么回事……_!_创新科技,银行
+6552479315353666051_!_104_!_news_finance_!_止损真的有必要吗，止损真的有用吗，只有1种情况才需要止损！_!_价值投资者,止损,投资者,基本面,股票
+6552485619187057160_!_104_!_news_finance_!_巴菲特股东大会上的珠宝店、家具店、牛排馆和乒乓球赛_!_股东大会,乒乓球赛,金融,巴菲特
+6552140014833107470_!_104_!_news_finance_!_钱程：如何认识K线反转形态并成功抓顶抄底_!_反转形态,形态,K线反转形态,K线反转,反转
+6552467490813772301_!_106_!_news_house_!_逾50万人新落户西安！这些城市都在推落户新政，楼市涨价在路上？_!_户籍,摇号,未央区,落户,西安市
+6552251859803832583_!_106_!_news_house_!_徐州与一线城市的差距主要在什么地方？_!_
+6552397213673193741_!_106_!_news_house_!_多年前，把宅基地卖给了城里人，现在后悔了，能要回来吗？_!_
+6552276258250555656_!_106_!_news_house_!_龙岩碧桂园小区环境，物业服务怎样？_!_
+6552252003861398019_!_106_!_news_house_!_没买房和计划买房的注意啦！_!_学区房,买房
+6552255307794350339_!_106_!_news_house_!_错过了楼市，错过了比特币，下一个风口会是什么？_!_
+6552357859345039624_!_106_!_news_house_!_郑州和西宁哪个城市更发达呢？_!_
+6552446848303366408_!_106_!_news_house_!_想在西安买房投资，哪个区域比较好，最好有具体楼盘？_!_
+6552460260261495309_!_106_!_news_house_!_张家口楼盘活力城超5A写字楼助你赢天下！_!_张家口,活力城,商业街,Citywalk,写字楼
+6552257095838728718_!_106_!_news_house_!_四月武汉最受关注度的20个楼盘出炉！大光谷片区上榜最多！_!_片区,楼盘,光谷,房价,武汉
+6552264963820355853_!_106_!_news_house_!_买了高层西边户会不会夏天很热，会受到什么影响？_!_
+6552261241803375117_!_106_!_news_house_!_产业地产开始转热_!_云计算,碧桂园,产业地产,越秀地产,房企
+6552459357315924487_!_106_!_news_house_!_高房价真能控制人口素质吗？_!_文化素质低,任志强,低素质低能力,人口素质,高房价
+6552444782453457155_!_106_!_news_house_!_“涨的不是房价，是货币发行量”这一说法是否有道理？_!_
+6552405080279089415_!_106_!_news_house_!_“现在抓紧买房，十年以后买不起”，对于这种说法你怎么看？_!_
+6552433520323592452_!_106_!_news_house_!_为什么很多企业家都说“未来投资房子是最大的错误”？_!_
+6552303770066223624_!_106_!_news_house_!_连跌12个月，厦门各区二手房成交价大曝光_!_温哥华,厦门房价,漳州港,高房价,成交价
+6552297600022741517_!_107_!_news_car_!_最牛国产豪车，比奥迪还漂亮，上市仅5分钟卖出8000+_!_吉利汽车,中控台,领克01,上海车展,沃尔沃
+6552301348484809229_!_107_!_news_car_!_车主随即抽中苏A88888，上牌花费不到一千，最终却放弃了这块牌照_!_张启发,5连号车牌,A88888,顺子号,货车,苏A88888
+6552438739451248904_!_107_!_news_car_!_吉利新帝豪与长安新逸动xt哪个好？_!_
+6552315844209148430_!_107_!_news_car_!_6座MPV车型宝骏360，起价5万，家用车市场的首选！_!_MPV,宝骏,家用车,宝骏汽车,宝骏730,宝骏360
+6552349115638874632_!_107_!_news_car_!_实体店根本买不到的9款车载小物，平时跑断腿都难买到，真好用_!_车载吸尘器,扶手箱,安全系数,水玻璃,遮阳板
+6552288683435753988_!_107_!_news_car_!_售价超过600万的库里南还不算贵？这几款车恐怕一般土豪也吃不消_!_柯尼塞格,CCXR,库里南,劳斯莱斯,Trevita,兰博基尼,老爷车,Veneno
+6552413513128624648_!_107_!_news_car_!_北京现代将推4款电动车 插电索纳塔7月上市_!_传祺GS4,北京汽车,新能源,伊兰特,GS4,蒙迪欧,索纳塔,GS4EV,plug-in,广汽丰田,北京现代,广汽三菱,PHEV
+6552380138900685320_!_107_!_news_car_!_京A88888到底是谁的车？真相浮出水面，看图知道原来是他！_!_李晓华,C88888,法拉利,A88888,保时捷,玛莎拉蒂,车牌,A00001
+6552321223169147139_!_107_!_news_car_!_金旅海狮车怎么样？_!_
+6552360916585808131_!_107_!_news_car_!_车用润滑油变质标准是什么？怎么才能看出来润滑油变质了？_!_
+6552338739497009667_!_107_!_news_car_!_现代领动最高优惠 最低报价 可分期_!_领动,万现车分期免息1.6L自动,北京现代,发动机,朗动,双离合
+6552412510924833284_!_107_!_news_car_!_义乌造的“睡觉神器”，连去酒店开房都不需要了，老司机百用不厌_!_司机,便携式,全自动,清洁软胶
+6552354679890641412_!_107_!_news_car_!_宝马汽车14款首发车型，最重要的还是这2款……_!_新时代,新能源,风向标,北京车展,BMW X3
+6552134601261711619_!_107_!_news_car_!_宝马旅行车和奥迪AllLOAD旅行车保养会很麻烦吗？_!_
+6552315874391360008_!_107_!_news_car_!_难怪老公跑长途从不喊累，原来有了这几款神器，简单实用_!_美丽演绎,家居服,活性炭,多功能,货到付款
+6552310520739791118_!_107_!_news_car_!_如果给你三十万，你会选择什么车？_!_
+6552319879091847683_!_107_!_news_car_!_战斗形态！新款野马5月底上市，或搭载10速变速箱！_!_极品飞车,速度与激情,福特野马,福特F-150,肌肉车,F-150,P-51,Mustang GT,Mustang
+6552365549030998541_!_107_!_news_car_!_昆明附近二十年的老牌烧烤，每天夜色降临，已停满200辆汽车_!_烧烤,昆明,安宁市
+6552454034299552004_!_107_!_news_car_!_宝马3系到底多靠谱？_!_
+6552164346267435523_!_107_!_news_car_!_夏季你是不是这样洗车？难怪新车越洗越旧！全搞错了，快来看看吧_!_冷胀热缩,洗衣粉,洗车
+6552276382171267597_!_108_!_news_edu_!_小学苏教版语文五年级下册每课一练生字词语句子课文内容专项练习_!_课课练,苏教版
+6552310212198400260_!_108_!_news_edu_!_铁路类大专哪家强？_!_
+6552439161507283460_!_108_!_news_edu_!_黄埔一期入学纪念日，回看黄埔军校教学水平究竟怎样_!_黄埔,高级将领,黄埔军校,军校,军事教育
+6552387129379389710_!_108_!_news_edu_!_职称评审期间，可以换工作吗？_!_
+6552495117968605709_!_108_!_news_edu_!_How to say this这样的中式英语，你说错了几个？_!_中式英语,Chinglish,free time,What
+6552402966983213315_!_108_!_news_edu_!_学校宿舍一般几人一个比较好？_!_
+6552359057762550029_!_108_!_news_edu_!_现代家庭教育中，是否存在所谓“知识倒挂”现象，你怎么看？_!_
+6552238249505456643_!_108_!_news_edu_!_语文中考复习：三看_!_课外阅读,经典考题,准确理解,语文,文言文
+6552294989118832909_!_108_!_news_edu_!_流浪狗扰乱正常教学秩序，破坏学生们生活，该怎么办？_!_
+6552460655234908680_!_108_!_news_edu_!_焦虑的社会，焦虑的中国式教育环境_!_逆反心理,教师
+6552273831245906190_!_108_!_news_edu_!_上专科选什么专业好？_!_
+6552421090184921603_!_108_!_news_edu_!_开封电子科技学校2018届学前教育专业毕业汇演精彩纷呈_!_学前教育,黄河颂，民族魂,舞起幸福鼓,开封电子科技学校,祝福祖国,军魂,最美宫灯红,黎乡笠影
+6552338426945864205_!_108_!_news_edu_!_千应助学——人才扶持培养政策实施方案_!_寒暑假,应助学,大学生,应学子,应网络
+6552277538519908615_!_108_!_news_edu_!_山西很多大学生又开始因为考研还是就业纠结，本科生毕业该如何抉择？_!_
+6552275782205440270_!_108_!_news_edu_!_女生初中毕业适合学什么专业？_!_
+6552428067015360776_!_108_!_news_edu_!_宜春发展前景怎样？_!_
+6552395158602646020_!_108_!_news_edu_!_西安交通大学筑梦社微宣讲团赴延安宣讲_!_延安大学,西安交通大学,西安交大,马克思主义,延安精神
+6552296122549797123_!_108_!_news_edu_!_西安航空学院和西藏民族学院哪个好？_!_
+6552493851926331656_!_108_!_news_edu_!_大学生是考研好，还是考公务员好？_!_
+6552426976999637251_!_109_!_news_tech_!_你愿意把十万元人民币存入支付宝，还是存入微信钱包？_!_
+6551234456315756814_!_109_!_news_tech_!_中国有没有能力自主研发高端芯片？_!_
+6552302164776059150_!_109_!_news_tech_!_有什么好的餐饮开业的营销方案？_!_
+6552397518796226820_!_109_!_news_tech_!_预算3000元，什么牌子的平板电视比较好？_!_
+6552293004982354435_!_109_!_news_tech_!_GoogleI/O2018还有两天就要来临，我们期待着什么？_!_Fitbit,Google,谷歌,Android Wear,Mirage
+6552137741323207176_!_109_!_news_tech_!_手机行业总体疲软的态势下，苹果为什么实现了逆势增长？_!_国产手机,苹果,市场需求,高端市场,市场份额,手机市场
+6552366757326094605_!_109_!_news_tech_!_2018互联网创业可行吗？_!_
+6552276587641831943_!_109_!_news_tech_!_支付宝里可以直接查银行卡余额了！_!_支付宝,兴业银行,余额
+6552302439351976206_!_109_!_news_tech_!_国内程序员那么多，为什么不开发一个属于自己的编程语言？_!_
+6552474971157823752_!_109_!_news_tech_!_Minecraft1.1.2版本中如何让实体方块能穿透（玩家能穿过）？_!_
+6552181511678853384_!_109_!_news_tech_!_如何看待微信小游戏的爆红，是真机遇还是伪风口？_!_
+6552274582991012104_!_109_!_news_tech_!_谁有好的微信头像可以分享？_!_
+6552474837011398916_!_109_!_news_tech_!_3000到5000的手机，有什么好的推荐？_!_
+6552387304806154765_!_109_!_news_tech_!_区块链人才紧缺，全球顶尖高校纷纷开课 中国八家高校搭上专车！_!_区块链实验与工程实践,比特币,数字货币,区块链,爱丁堡大学,全球高校区块链课程报告,区块链原理与技术,麻省理工学院,赛博智能经济与区块链
+6552383158208168452_!_109_!_news_tech_!_马云告诫以色列大学生：别介意出身，重要的是你做什么_!_马云,Einstein,特拉维夫,以色列,小企业
+6552377939386695943_!_109_!_news_tech_!_做算法工程师是什么样的体验？_!_
+6552381648715907341_!_109_!_news_tech_!_红信APP真的可以领红包？_!_
+6552335303783219469_!_109_!_news_tech_!_现在手机网络发展是不是过快了？_!_
+6552464336957210883_!_109_!_news_tech_!_如果安卓也像ios一样，不理智消费，可以申请退款，中国会有多少家公司要倒闭？_!_
+6552410257774084356_!_109_!_news_tech_!_GPP排他性协议引发巨大争议，Nvidia该怎么收场？_!_
+6552304092025192711_!_110_!_news_military_!_为什么美国总统出国访问都要空运专车？_!_
+6552268410204455171_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552448166224986372_!_110_!_news_military_!_中国辽宁航母战斗群的战斗力如何？_!_
+6552285683967328781_!_110_!_news_military_!_美将用最先进导弹打击东风快递，性能超群，专家：难以探测_!_精准打击,国家利益,AGM-158,涡轮喷气发动机,固体燃料,巡航导弹
+6552277385767551502_!_110_!_news_military_!_无人机还敢乱飞？我军两架战机前往拦截！四人被抓！_!_玩无人机,范各庄,河北省,架无人机,唐山市
+6552415300938105101_!_110_!_news_military_!_北约16国在俄家门口举行军演，调动大批坦克战机，俄军将如何应对？_!_
+6552374393031360771_!_110_!_news_military_!_四野在东北的武器装备是哪里来的？_!_
+6552358243228713486_!_110_!_news_military_!_国产航母尚未海试，其超强舰载机已试飞，俄：核心源自我们_!_歼-15,电子战机,WS-10,AL-31,电子战,歼-11,舰载机
+6552310176702005774_!_110_!_news_military_!_征兵季！盘点军营5大“上古神器”, 看到第一个我就跪了_!_上古神器,一甲子,部队,工兵锹
+6552349627562066184_!_110_!_news_military_!_五四运动中遭世人唾骂的卖国贼曹、章、陆三人，最终结局如何？_!_
+6550640121702514952_!_110_!_news_military_!_机械扫描雷达与相控阵雷达主要差异是什么？_!_
+6552309649352163843_!_110_!_news_military_!_叛军准备反扑，却遭遇俄军撒下白磷弹，俄罗斯斩获重大成果_!_白磷弹,伊德利卜,俄罗斯,哈马,叛军
+6552260339067519246_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552277883203617294_!_110_!_news_military_!_全球最顶尖的 无人机_!_以色列,捕食者,攻击机,星球大战,RQ-4A,无人机,全球鹰
+6552259391855264263_!_110_!_news_military_!_军装承载荣耀 时刻践行标准_!_军人,中国人民解放军内务条令,穿军装,军装,空军
+6552452563378110990_!_110_!_news_military_!_美媒：俄罗斯的苏-27战机令北约害怕_!_苏-27,F-15,俄罗斯,中短程,防空导弹
+6552278560344637956_!_110_!_news_military_!_盘点全球十大速度最快的 飞机_!_苏霍伊,F-15,Su-27,YF-12,X-2,战斗机,苏-27,米高扬,SR-71,侦察机,XB-70,A-12,F-111
+6552460632262705415_!_110_!_news_military_!_历史上的十大军事人才有谁？_!_
+6552380095414141454_!_110_!_news_military_!_征兵高清海报来了！想去部队当兵的你还在犹豫什么？_!_部队
+6552403672658084110_!_110_!_news_military_!_军事上常用的炸药，除了TNT，还有哪些类型？_!_
+6552291440674734595_!_112_!_news_travel_!_因抖音爆红的六大景区，你去过几个？网友：越红越不要去_!_西塘古镇,李子坝站,鼓浪屿,小宝贝,鬼哭狼嚎,丽江古城
+6552453140338180359_!_112_!_news_travel_!_四川游客张某带着六名女子在三亚沙滩烧烤被撒沙，这事你怎么看？_!_
+6551753955225895437_!_112_!_news_travel_!_塞班军舰岛：一个无数比基尼美女扎堆的海岛，搞不好还可以艳遇哦_!_塞班岛,军舰岛,塞班,北马里亚纳群岛,太平洋
+6552316376848007688_!_112_!_news_travel_!_孟津隐藏的这处美景，你来过吗？趁着周末约起来！_!_孟津,人工湖,南岭,生态水系
+6552373137676173838_!_112_!_news_travel_!_妙龄少女过玻璃栈道，路人眼睛都看直了！_!_玻璃,栈道,玻璃栈道
+6552331576603050248_!_112_!_news_travel_!_自驾游西藏无人区加油问题怎么解决？_!_
+6552429191420182797_!_112_!_news_travel_!_《雍正王朝》四阿哥胤禛和十三阿哥胤祥，办差路过江夏镇为什么必须下马步行通过？_!_
+6552281983114805764_!_112_!_news_travel_!_岳阳市温暖湿润，季节性强，热量丰富，严寒期短、无霜期长_!_
+6552069923793797640_!_112_!_news_travel_!_说动就动，说拍就拍_!_
+6552471885643251982_!_112_!_news_travel_!_盐城哪里最好玩？_!_
+6552282462771216909_!_112_!_news_travel_!_世界上岛屿最多的湖-杭州千岛湖_!_
+6552320160366067981_!_112_!_news_travel_!_佛山有哪些地方可以游水？_!_
+6552416532247347725_!_112_!_news_travel_!_广安十六景 景景皆有情_!_儒林,广安志,通志,归县志,广安州新志.山川志,流觞曲水,渠江,川北道,旧州志,广安,甘泉寺,广安十六景,广安州志,老县城
+6552298322860704269_!_112_!_news_travel_!_水墨石林——甘肃景泰石林美景_!_甘肃,石林,东方IC,水墨石林,甘肃景泰石林美景
+6552382066023989511_!_112_!_news_travel_!_贵州哪里有比较大的服装批发商铺？_!_
+6552491027570098691_!_112_!_news_travel_!_昆明这个地方的蓝花楹美得不像样，一生中一定要这样欣赏一次_!_蓝花楹,昆明,盘龙江,林荫大道
+6552443242627990019_!_112_!_news_travel_!_2018金秀瑶族自治县圣堂山杜鹃花旅游文化节推介会在南宁召开_!_金秀瑶族自治县圣堂山杜鹃花旅游文化节,金秀,金秀瑶族自治县,金秀瑶族自治县圣堂山杜鹃花旅游,圣堂山杜鹃花旅游文化节
+6552290735956165128_!_112_!_news_travel_!_专业探险队深入洞穴探洞，发现里面的宝贝可真不少！_!_麦穗鱼,探险队,钟乳石,小黄哥,筠连县
+6477778335580553486_!_113_!_news_world_!_在广西东兴可以开车去越南芒街吗？_!_芒街市,越南,中国自驾游汽车,广宁省,芒街
+6524226828394037774_!_113_!_news_world_!_这条鱼被“通缉”了，他究竟做了什么？_!_亚洲鲤鱼,澳洲,鲤鱼,美国
+6523523913643721229_!_113_!_news_world_!_世界犬种介绍之30：犬中帝王——阿富汗猎犬_!_伴侣犬,犬种,阿富汗猎犬,阿富汗
+6549792116325745166_!_113_!_news_world_!_第一位有中国国籍的外国人，留中国五十多年，死前留下这样的话！_!_周苏菲,马海德,毛主席,解放战争,翻译官
+6531295423976440077_!_113_!_news_world_!_萨达姆死后，他的后代现在在何方？_!_
+6548382485170881027_!_113_!_news_world_!_中国与越南边界_!_加油机,越南
+6547611371964465677_!_113_!_news_world_!_女儿美到路人自动献花，爸爸辞职当“保镖”_!_风云人物,伊朗,马蒂
+6538385726214504963_!_113_!_news_world_!_中国游客在巴基斯坦很受欢迎，如同贵宾，但这个方面很严格_!_中国游客,巴基斯坦,卡拉奇,巴铁,驴行天地间,联合国
+6540502795273896195_!_113_!_news_world_!_为什么说去泰国旅游，千万不能去捡海边的贝壳？_!_
+6550946277981422094_!_113_!_news_world_!_CR929大飞机俄总师：没有我们帮助，中国造不出质量高的大客机_!_大型宽体客机,CR929,宽体客机,俄罗斯,客机
+6525036281347441160_!_113_!_news_world_!_为何日本自古以来一直想吞并中国？说出来你可别不信！_!_日本,吞并中国,火山地震,西伯利亚,白江口之战
+6552778284403261959_!_113_!_news_world_!_卡扎菲和他的美女保镖们，他们之间真像外界传说的那样不堪吗？_!_美女保镖,女保镖,卡嘉莉,卡扎菲,利比亚
+6516634794988667140_!_113_!_news_world_!_多少只猛犬才能打的过东北虎，甚至能杀死老虎？_!_高加索,老虎,斗牛犬,高加索犬,猛犬
+6550863425008828941_!_113_!_news_world_!_安倍最想销毁的瞬间照，十分尴尬，颜面尽失_!_丰田章男,安倍,安倍晋三,奥巴马,朴槿惠
+6551947069249028621_!_113_!_news_world_!_老外挑战100小时不睡觉，看完后我默默的盖上了被子_!_保龄球,日出而息,仓鼠
+6552127971371516429_!_115_!_news_agriculture_!_乡村全面振兴将在大中城市率先实现_!_大中城市,菜篮子,质量兴农倡议书,韩长赋,农村,农业
+6552363872114704909_!_115_!_news_agriculture_!_农村大集上54岁腼腆大哥靠竹丝大扫帚老手艺30年，年赚一万元_!_
+6552259842835218958_!_115_!_news_agriculture_!_被很多农民嫌弃的野生植物，40元至50元一斤，快看看你家地里有吗_!_半夏,野生植物,农村
+6552485759624937991_!_115_!_news_agriculture_!_下个周末崂山北宅樱桃将迎来集中上市期 结伴去体验采摘乐趣吧_!_樱桃,崂山,北宅,采摘期,北宅樱桃
+6552367679045370115_!_115_!_news_agriculture_!_鸡农药中毒后，有什么好的解救方法？_!_
+6552447961404539149_!_115_!_news_agriculture_!_农村集市上的垃圾桶真的可以卫星定位吗？_!_
+6552019496138179079_!_115_!_news_agriculture_!_玉米大小苗严重影响产量，及时采取这些措施，减少损失多赚钱！_!_小苗,播种,播种时,玉米出苗,发芽率
+6552379149976076558_!_115_!_news_agriculture_!_一些农民收割油菜后，为什么要把它放在家里堆上几天才去翻晒求籽？_!_
+6552411968269976078_!_115_!_news_agriculture_!_农村常见这种植物被称作生命之花，将它放在卧室有利于夫妻和睦！_!_绿萝花,黄金葛,糖尿病,绿萝,天南星科
+6552439289043484936_!_115_!_news_agriculture_!_为什么有的果树旺长却不结果？如何改造成高产树？_!_
+6552299045006606599_!_115_!_news_agriculture_!_西红柿苗期冲施噻唑膦后发黄怎么回事？_!_
+6552392450013397507_!_115_!_news_agriculture_!_正是这几点原因，使得美国牛肉如此便宜_!_农业补贴,畜牧业,农业,肉牛,瘦肉精
+6552418376650588419_!_115_!_news_agriculture_!_英国乡村风光最好的只有Cotswolds吗？_!_
+6552351210970546692_!_115_!_news_agriculture_!_村民地下挖出“人形大人参” 给专家鉴定后却是另外回事！_!_中药材,天花粉,人参
+6552290462646927876_!_115_!_news_agriculture_!_县区丨新田：小辣椒成贫困村致富“火红产业”_!_扶贫队,贫困户,郑永发,新田二中,土鸡
+6552250748787229192_!_115_!_news_agriculture_!_东北多地要求多种大豆 要用中国产替代进口？_!_政治,农业委员会,大豆,长春市,农业生产
+6552397180261368328_!_115_!_news_agriculture_!_农村的这些花花草草，你能随便叫出名字？_!_雌雄异株,喇叭花,田旋花,牛筋草,泥胡菜
+6552414050020491784_!_115_!_news_agriculture_!_立夏狂风暴雨，西安市民都不淡定了，你怎么看？_!_农作物病虫害,立夏,长江中下游,皮肤病
+6552430911344869902_!_115_!_news_agriculture_!_警惕农业天价传销_!_大善人,农业
+6552256975818719752_!_115_!_news_agriculture_!_养猪知识：母猪合圈方法 检验母猪是否怀孕方法_!_玻璃瓶,尿液,怀孕,母猪,养猪
+6552266663226507523_!_116_!_news_game_!_想玩恐怖游戏又害怕怎么办？_!_
+6552128557970096654_!_116_!_news_game_!_王者荣耀：英雄池中必须有这两名英雄，才是一名好的辅助玩家_!_打野,姜子牙,BUFF,鬼谷子,打野英雄
+6551608901404459533_!_116_!_news_game_!_VR 3.5渲染器大图渲染设置，再也不担心不会设置了_!_图渲染设置,VR 3.5版本发布,设置,渲染,VR 3.5
+6552430319767650820_!_116_!_news_game_!_绝地求生：游戏中成就系统37项你完成了几个？_!_十字弩,游戏
+6552299649116406276_!_116_!_news_game_!_麦当劳开王者荣耀主题店了？可以开黑吗？_!_麦当劳,开王者
+6552410423121936900_!_116_!_news_game_!_王者荣耀：有这个英雄在，辅助位都能变成团战中心_!_博学者,钟馗,X10,敌方英雄,黄忠
+6552385005778108685_!_116_!_news_game_!_王者荣耀里钟馗的钩子和百里玄策的钩子哪个更可怕，你更怕哪个？_!_
+6552411574026371592_!_116_!_news_game_!_回顾LSPL比赛！让帝被Khan爆锤，才明白FW前教练之前说的话_!_Khan,让帝,RNG,MSI
+6552436745860809229_!_116_!_news_game_!_UZI直播 网友：震惊，韩服前十被国服黄金吊打_!_uzi,游戏,国服,韩服
+6551621685777269255_!_116_!_news_game_!_《我的世界》：它是最可怜的生物，想要靠近玩家，却又让玩家害怕_!_Creeper,沙盒游戏,苦力怕,爬行者,我的世界
+6552418687431737860_!_116_!_news_game_!_《业余游玩》steam周销量排行榜_!_Saga,不列颠,反恐精英,Steam,野性的呼唤
+6552239764798439684_!_116_!_news_game_!_《英雄联盟》RNG战队北京主场招聘赛事OB,要求段位电一大师以上，你怎么看？_!_
+6552366326298444296_!_116_!_news_game_!_绝地求生12轮更新新增内容讲解：新配件与倍镜的抉择_!_倍镜,握把,后坐力,SKS,突击步枪
+6552481748427473416_!_116_!_news_game_!_天龙八部手游：华山论剑特殊战法就在这里！_!_呆若木鸡,华山论剑,华山,隐身,输出
+6552390207054807560_!_116_!_news_game_!_LOL如何不出高地赢下比赛？这件神装必备！_!_lol,防御塔,超级兵,高地,ADC
+6552359742533009668_!_116_!_news_game_!_如果你喜欢《我的世界》，但是你妈妈又说《迷你世界》好，你会怎么办？_!_
+6552337065843884551_!_116_!_news_game_!_现在想要脱坑都这么难，老马硬生生拉玩家回坑，这代价有点大_!_金闪闪,DNF,老马,圣耀太刀,脱坑
+6552329043792888077_!_116_!_news_game_!_lol哪个英雄最吃操作？_!_
+6552490727174046212_!_116_!_news_game_!_LOL：被安排？抛弃了女朋友和小明双排的Uzi首秀屡次却被黄金单杀_!_德莱文,直播间,Uzi,MSI,艾欧尼亚
+6552489906545885704_!_116_!_news_game_!_路人局无解的阿珂，为何在KPL无人问津？梦泪亲自告诉你答案！_!_阿珂,路人局,雅典娜,李元芳,KPL
+6552329846385541646_!_101_!_news_culture_!_雕工丑 VS 朴拙美，“另类”雕刻引争议_!_
+6552421221621826056_!_101_!_news_culture_!_书法学习之77_!_书法,癞蛤蟆
+6552427447701209352_!_101_!_news_culture_!_上联：密密麻麻雨，轰轰隆隆雷，下联怎么对？_!_
+6552382368680772103_!_101_!_news_culture_!_看完邱会作墓地，再看看黄永胜墓，意义深远，很少人去_!_咸宁市,黄永胜,邱会作,湖北省,兴国县
+6552260691019956743_!_101_!_news_culture_!_“天虹七彩曜变”宁波茶博会再续天目传奇 探寻建盏复兴之路_!_天虹七彩曜变,茶文化,建盏,高丽图经,茶博会,宁波
+6552419274521051656_!_101_!_news_culture_!_古诗词美文书法欣赏——游子春衫已试单，桃花飞尽野梅酸_!_书法,野梅酸,古诗词,桃花
+6552478671792242952_!_101_!_news_culture_!_《西游记》中的各路神仙，腾云驾雾动辄十万八千里，为什么没有发现地球是圆的呢？_!_
+6552390932254163214_!_101_!_news_culture_!_五胡乱华时期到底有多黑暗？_!_
+6552416440916378125_!_101_!_news_culture_!_女子出一上联：一杯清茶，解解解元之渴！解缙至死都没对出下联_!_李尚书,永乐大典,解缙
+6552157096303919630_!_101_!_news_culture_!_萧山古墓群损毁严重 文物贩子每块200元兜售汉墓砖_!_古墓群,文物贩子,浙江,萧山,文物
+6552489697216561678_!_101_!_news_culture_!_辛弃疾和党怀英的分叉路口_!_稼轩记,张安国,周易,辛弃疾,党怀英,安抚使
+6552374317064126728_!_101_!_news_culture_!_时光飞逝，无计留春住。怎么对下联？_!_
+6552278890033709316_!_101_!_news_culture_!_林徽因为什么对金岳霖产生互爱？_!_
+6552332223800934660_!_101_!_news_culture_!_在中国的历史上，你认为谁最伟大？_!_
+6552278735209366023_!_101_!_news_culture_!_管窥西泠印社前任副社长 大师方去疾篆刻艺术 01_!_明清篆刻流派印谱,中国书法家协会,青田石,方去疾,西泠印社,白文
+6552470018519466253_!_101_!_news_culture_!_生肖是按阴历还是阳历来算？_!_
+6552236338492473608_!_101_!_news_culture_!_马克思是世界上最伟大的哲学家吗？_!_
+6552254870060007684_!_101_!_news_culture_!_上联：春夏秋冬交好运，如何对下联？_!_
+6552323220500906247_!_101_!_news_culture_!_朱元璋为何要将沈万三满门抄斩？_!_
+6544495786992861700_!_102_!_news_entertainment_!_原来，那只是她自己的遗憾，与对方无关_!_前任3,电影院,电影,后来
+6552486397532438792_!_102_!_news_entertainment_!_如何评价《权力的游戏》里的托曼国王？_!_
+6552310256754491911_!_102_!_news_entertainment_!_她曾被誉为“王昭君”，如今整得亲妈认不出，穿衣大胆不输范冰冰_!_女星,回家的诱惑,性感,李彩桦,范冰冰,王昭君
+6552279959933551107_!_102_!_news_entertainment_!_李梦：出席活动，网友：这穿着打扮真是让人匪夷所思一饱眼福_!_匪夷所思
+6552464403273351693_!_102_!_news_entertainment_!_张杰写给谢娜的信，女人看了都会感动_!_谢娜,女人,张杰,娱乐圈,关爱和
+6552354880877494788_!_102_!_news_entertainment_!_测试发文1525589004.3_!_
+6552473995462050051_!_102_!_news_entertainment_!_你曾经崇拜过哪些现在看来很丑的偶像？_!_
+6552441196021547534_!_102_!_news_entertainment_!_刘诗诗：出席活动，网友：天鹅颈只服刘诗诗，美呆了！_!_天鹅颈,鹅颈,刘诗诗,服刘诗诗
+6552383525306237453_!_102_!_news_entertainment_!_杨幂首次回应，一开口粉丝就炸了_!_刘恺威,上升期,杨幂,网友粉丝,仙剑奇侠传三,仙剑奇侠传3,胡歌
+6552306969380651277_!_102_!_news_entertainment_!_行云流水，句如晨风迎朝霞，如何接句？_!_
+6552392738531181070_!_102_!_news_entertainment_!_她曾因一部剧走红，但为了姐弟恋淡出荧幕，如今丈夫却红得了不起_!_姐弟恋,男二号,李佳璘,小李琳,李琳
+6552438323195937287_!_102_!_news_entertainment_!_谢娜真的可以退圈了，粉丝过亿被催娃，产后迅速回归却被网友嫌_!_快乐大本营,谢娜,蔡少芬,Anglebaby,粉丝,佟丽娅
+6552314755082617351_!_102_!_news_entertainment_!_李小璐出席某活动，网友：你不用撩裙子也有人找你做头发_!_李小璐李小璐,李小璐
+6552462446152384772_!_102_!_news_entertainment_!_如何评价演员小张铎的演技？_!_
+6552484485177278983_!_102_!_news_entertainment_!_拒绝雷同！《幻乐之城》全原创模式 呈现音乐加表演顶级大秀_!_梁翘柏,音乐节,台前幕后,幻乐之城,湖南卫视
+6552353716257686020_!_102_!_news_entertainment_!_陈立农粉丝见面会 演唱浓糖版的《纸短情长》！真的太暖心了！_!_陈立农,见面会,粉丝,纸短情长
+6552368456723857924_!_102_!_news_entertainment_!_张靓颖间歇性失聪_!_The Diva Dance,终于等到你,如果爱下去,中世纪,演唱会,Fighting Shadows,顺流而下,张靓颖,如果这就是爱情,凯迪拉克,珍相
+6552286644534247943_!_102_!_news_entertainment_!_光头后的男明星！李晨可以拍科幻片！谢霆锋颓废感刘德华更加帅气_!_刘德华,谢霆锋,吴亦凡,李晨,彭于晏
+6552298830040137998_!_103_!_news_sports_!_瑞典世乒赛中国女队3-1日本女队卫冕，如何看待刘诗雯的表现？_!_
+6552231114260349187_!_103_!_news_sports_!_哈登和保罗激烈争吵，保罗伸手还被哈登打飞，安德森全程看呆，最后相互不理，怎么看？_!_
+6552382213000790535_!_103_!_news_sports_!_维特尔：近几年莱科宁的状态被人们曲解了_!_维特尔,法拉利,Ferrari,赛车,莱科宁
+6552454358049489166_!_103_!_news_sports_!_NBA的三分谁最厉害？_!_
+6552339325781017092_!_103_!_news_sports_!_八年抗詹，何时是头！明年詹姆斯该对东部球队进行九年义务教育_!_步行者,NBA,格兰杰,凯尔特人,骑士
+6552425199474573831_!_103_!_news_sports_!_北控女足主场惜败_!_北控,中国女足,北京女足,先农坛,先农坛体育场
+6552356189798138372_!_103_!_news_sports_!_2018陆家嘴金融城足球联赛昨天拉开帷幕_!_海尔集团,足球联赛,陆家嘴金融城足球联赛,金融城,循环制
+6552300785131061767_!_103_!_news_sports_!_骑士再胜猛龙拿到赛场，场边泰伦卢采访通过乐福花式盛赞詹姆斯_!_猛龙,泰伦卢,常规赛,詹姆斯,骑士
+6552331919692923395_!_103_!_news_sports_!_泰伦-卢：当詹姆斯全力冲刺时，没人能防得住他_!_勒布朗,泰伦,詹姆斯,卢说道,季后赛
+6552302888457077252_!_103_!_news_sports_!_詹姆斯又压哨绝杀！联盟第一人毫无争议？_!_系列赛,骑士,德罗赞,乐福,詹姆斯
+6552367462975799821_!_103_!_news_sports_!_裕安区平桥乡助力国际马拉松赛_!_国际马拉松赛,平桥乡,马拉松,裕安区
+6552398193680712206_!_103_!_news_sports_!_世界杯第一战队卷土重来！集齐六大巨星强攻大力神杯_!_热苏斯,内马尔,库蒂尼奥,巴西足球,巴西
+6552459119209480462_!_103_!_news_sports_!_奥拉朱旺远赴盐湖城为火箭队加油助威，你如何评价？_!_
+6552290881989247245_!_103_!_news_sports_!_火箭被爵士打出9-0反击，保罗，哈登为此争吵，从中我们能够看出什么？_!_
+6552468346367902215_!_103_!_news_sports_!_遭遇3连败，东部2队面临生死一战！坐拥主场优势，大帝有望破局？_!_德罗赞,绿军,骑士队,猛龙队,76人队
+6552269235421184259_!_103_!_news_sports_!_休斯顿火箭队哈登和雷霆队威少，还有金州勇士队杜兰特，谁的实力更强？_!_
+6552327527245807879_!_103_!_news_sports_!_丢一场赢一场，今天刘诗雯在世乒赛女团决赛的表现值不值得表扬？_!_
+6552360987167555854_!_103_!_news_sports_!_勒布朗詹姆斯和骑士如何改变季后赛命运？_!_
+6552332540181479939_!_103_!_news_sports_!_「滚球体育APP」北京人和vs山东鲁能直播分析和预测推荐_!_鲁能,山东鲁能,中超,点球大战,北京人和
+6552382748558885380_!_104_!_news_finance_!_妖股第二次砸盘，欢乐海岸还会来反包吗？_!_国泰君安,GQY视讯,太平南路,华锋股份,永和智控,欢乐海岸,金刚玻璃
+6552384654672921101_!_104_!_news_finance_!_华峰股份：3月内不再筹划除购买华创股权之外重大事项_!_北京理工,华峰,电动车,股票,股权
+6552486114484027656_!_104_!_news_finance_!_分红后要除权，股价相应下降，股东也没多赚到钱，意义是什么？_!_
+6552139359036899588_!_104_!_news_finance_!_期货交易到底难在哪里？_!_
+6552371940714086670_!_104_!_news_finance_!_纸质水瓶面世了，塑料瓶子在饮料包装市场的命运将如何？_!_
+6552259222589932046_!_104_!_news_finance_!_零售企业开发自有品牌的三大痛点_!_零售企业,零售商,商品,领导品牌,供应商,大润发,沃尔玛
+6552131889010835976_!_104_!_news_finance_!_一个创业成功的项目，是这样做的_!_选择项目,创业者
+6552393633184612621_!_104_!_news_finance_!_如何应对加密货币的崩溃？_!_
+6552437257398124804_!_104_!_news_finance_!_中金e配股票配资是实盘操作吗？_!_
+6548318357102264584_!_104_!_news_finance_!_分红险、万能险、投连险有什么区别？_!_
+6552365860193829384_!_104_!_news_finance_!_为何有的人信用卡能做到快速提额？因为是这个原因_!_云闪付,账单日,还款日,信用卡,提额
+6552308199360299268_!_104_!_news_finance_!_18年大蒜的市场行情怎么样，价格有没有可能提高？_!_
+6552373618389549325_!_104_!_news_finance_!_关于均线系统，有什么独特的、有效的见解跟设置用法吗？_!_
+6552401133837484295_!_104_!_news_finance_!_股票增发和股票回购是正好相反的两种操作吗？_!_
+6552351208802091277_!_104_!_news_finance_!_假设一只股票开盘谁都不买，那是不是只要在涨停板挂一手买单，就能涨停？_!_
+6552240760916279556_!_104_!_news_finance_!_如何系统地学习股市？_!_
+6552302407072612611_!_104_!_news_finance_!_股市究竟应该如何去学？_!_
+6552371789803028739_!_104_!_news_finance_!_“国六”即将实施，常用的“国三”“国四”会怎样处理呢？_!_
+6552357318057525517_!_104_!_news_finance_!_买了太平人寿保险，得了病花了20万，保险公司说没有动大手术拒赔怎么办？_!_
+6552436591262957831_!_106_!_news_house_!_户口不在海南，在海南买房会限购吗？_!_
+6552322216602632712_!_106_!_news_house_!_日本新干线与中国高铁有啥不同_!_地震,新干线列车,日本,ATC,高速铁路,日本新干线
+6552278519773135112_!_106_!_news_house_!_在昆明月薪3500，怎么买房？_!_
+6551934749655433732_!_106_!_news_house_!_上海 4 月份新房供应量创 22 个月来新高！_!_成交量,虹桥商务区,新盘,供应量,新增供应
+6552316595534823683_!_106_!_news_house_!_月薪1.2万，房贷1500元，适合买什么车？_!_
+6552253114592788999_!_106_!_news_house_!_如何看待房子的涨跌？普通人买房有希望吗？_!_投资属性,王健林,房价,居住属性,城镇化水平
+6552430288884990222_!_106_!_news_house_!_红寺堡区有房屋中介信息部吗？具体在哪？_!_
+6552159413170340099_!_106_!_news_house_!_燕郊房价还会涨到调控前的水平吗？_!_
+6552288389494735363_!_106_!_news_house_!_房地产四大龙头企业销售榜单出来了，碧桂园强势登顶_!_买房,碧桂园,房价,房地产,销售额
+6552305678000587015_!_106_!_news_house_!_楼盘售罄后，销售员去哪儿了？_!_
+6552414014880612868_!_106_!_news_house_!_谁再劝你先别买房，就把这10条狠狠甩他脸上！_!_房产,买房,房价,高房价
+6552440644994859527_!_106_!_news_house_!_每日房源推荐丨又是一年丁香花开_!_郑州市,房源,河南省,WIFI,丁香花
+6552357854517395981_!_106_!_news_house_!_未来大部分房价会跌，但这15个城市或最稳定，看看有没有你家_!_买房,西双版纳,房价,香港大学,楼市
+6552400258872115725_!_106_!_news_house_!_「大事记」周巷安邦郡园最新消息_!_慈溪市,大事记,杭州湾新区,周巷,周巷安邦郡园最新消息
+6552105570726314499_!_106_!_news_house_!_抢房的新西安人_!_房价,二手房,新西安人,城六区,一手房
+6552480152494801412_!_106_!_news_house_!_二线城市房价借风助燃，你还在犹豫什么？_!_买房,安居客,外地人,房价,南京
+6552430622248272391_!_106_!_news_house_!_今年西安第18批备案价出炉 754套房源入市_!_房源,购房者,买房,楼盘,西安市
+6552404540606382350_!_106_!_news_house_!_在大连西山水库买房养老好不好？_!_
+6552362984545452548_!_107_!_news_car_!_德系血统，比朗逸高级，当年售价近20万，今跌至9万十分厚道_!_福克斯,卡罗拉,性价比,本田思域,速腾
+6551631623828275725_!_107_!_news_car_!_西安机动车保有量突破300万 交通治堵成焦点_!_
+6552322381728186632_!_107_!_news_car_!_汽车，国三，国二的车型有什么保值的？_!_
+6552178263840522766_!_107_!_news_car_!_别总想着换发动机！这几个关键部位不检查，难怪车子越开越耗油_!_清洗剂,润滑油,积碳,逆变器,发动机
+6552247922589696263_!_107_!_news_car_!_宝骏510跑工地怎么样？_!_
+6552474299855274243_!_107_!_news_car_!_奇瑞汽车使用的CVT变速箱质量如何？_!_
+6552441084968960526_!_107_!_news_car_!_紧急提醒！有工行、建行、农行卡的人，一定要看！_!_
+6552312890387333645_!_107_!_news_car_!_新一代唐只是个开始，全新秦这颜值实力，能让国产新能源市场翻天_!_新能源,全新秦,北京车展,中型车,比亚迪
+6552412512548028932_!_107_!_news_car_!_新一代车载逆变器来袭，你的移动发电站，出门用电就是这么简单_!_空气净化器,充气泵,逆变器,救生手环,除静电
+6552382962040570382_!_107_!_news_car_!_方向盘摆正时打死到底是多少圈？究竟是1.5圈还是2圈？_!_车标
+6552355715372024077_!_107_!_news_car_!_机油滤芯有一次性和可换纸滤芯的，为什么有设计区别？_!_
+6552489821401514504_!_107_!_news_car_!_传统弹弓早就过时了！当下特流行这强悍狠货，功能霸气更给力_!_牛津布,钥匙扣,过滤烟嘴,野餐垫,沙发床
+6552481129184625156_!_107_!_news_car_!_劲客试驾体验记：明明可以靠脸吃饭，却偏偏要靠实力！_!_劲客,Sport,V-Motion,R17,E70,发动机,CVT,CVT变速箱
+6552351399563231751_!_107_!_news_car_!_武威城区新增1条禁停路段！_!_停道路,中华人民共和国道路交通安全法,机动车
+6552396884789428750_!_107_!_news_car_!_世界上最贵的手动挡轿车 没有安全气囊ESP 比奔驰S级还贵_!_手动挡,摩根Roadster,Roadster,奔驰S级,保时捷911,手动变速箱,老爷车
+6552377815327572227_!_107_!_news_car_!_1.0T思域怎么样？适合家用吗？_!_
+6552354076338684174_!_107_!_news_car_!_预算12万，荣威rx5和长安cs75买哪个好？_!_
+6552459576795464196_!_107_!_news_car_!_这款最帅气的国产SUV一来，奥迪A7失去光芒，售价和油耗都低_!_SUV,马自达,奥迪A7,T90,启辰,启辰T90,奥迪,CX-4
+6552398557423337731_!_107_!_news_car_!_落地价十七八万，适合二十来岁年轻人开的中型车有哪些？_!_
+6552292435307790856_!_108_!_news_edu_!_北师大版一年级下学期数学期中试卷，家长可按答案评分标准改卷！_!_北师大
+6552344290637382148_!_108_!_news_edu_!_菲律宾学英语「宿务」C2 Ubec English Academy_!_ESL,C2,English,奥格登,选修课
+6552296735039816199_!_108_!_news_edu_!_教师资格证作文高分素材—倾听_!_林肯,卡耐基,苏霍姆林斯基,植物学家,季娜
+6552399443306480142_!_108_!_news_edu_!_家长速看！外地户口的小朋友在威海上小学都需要什么？_!_威海,房产,居住证,户口,户口本
+6552404956043804932_!_108_!_news_edu_!_英语六级很难吗？_!_
+6552141745272914190_!_108_!_news_edu_!_如果从江西高校中挑选5所很好的大学，你会如何挑选？_!_
+6552135762798706947_!_108_!_news_edu_!_毕业后怎么重新修读其他专业？_!_
+6552311300297327107_!_108_!_news_edu_!_董卿《朗读者》：一份给孩子最好的教育，为孩子收藏，作文满分！_!_董卿
+6552447779233333507_!_108_!_news_edu_!_作为一个高三学生的家长，为孩子的高考志愿填报，现在应该做什么？_!_
+6552322805701017860_!_108_!_news_edu_!_为什么要有中考？_!_
+6552401247188549902_!_108_!_news_edu_!_为什么1÷3再×3的结果不等于1？_!_
+6552461670998868484_!_108_!_news_edu_!_这是长沙最难脱单的一所学校，没有人不服！_!_女子学院,湖南女子学院,长沙
+6552308371377095171_!_108_!_news_edu_!_唐河县文峰街道人大代表、政协委员检查指导辖区教育工作_!_大班额,文峰街道,李延宗,街道教育,人大代表
+6552336808410087943_!_108_!_news_edu_!_小学生“奇葩”试题，家长被“虐哭”，网友：能做出一道就是人才_!_小学生,大学生,神题
+6552328747146543620_!_108_!_news_edu_!_那个给“二本师范生”讲课的奥数满分天才，在自白书里这样谈起“坠落”_!_小李飞刀,付云皓,张益唐,奥数天才坠落之后,奥数,人物,付云皓，一等牛人,师范生,IMO,星际争霸,广州大学,小学应用题大全,数学竞赛,魔兽世界,杨建生
+6552267697516708099_!_108_!_news_edu_!_教师资格证面试试讲如何分配时间？_!_
+6552362707176129028_!_108_!_news_edu_!_ESI中国最新十大高校排行，第一名你也许不知道_!_中国大学校长联谊会,中华人民共和国教育部,山东大学,环太平洋大学联盟,九校联盟
+6552299887134769668_!_108_!_news_edu_!_县区丨东安：扎实推进阳光体育 有效增强学生体质_!_东安县,东安武术,体育运动,中学生,体育活动,排球赛
+6552299262829396231_!_108_!_news_edu_!_西安的高校有哪些“最”？_!_
+6552369686204383495_!_109_!_news_tech_!_小米7支持5G吗？_!_
+6552487434708320519_!_109_!_news_tech_!_被拍拍贷追债什么感觉？_!_
+6552485090486649101_!_109_!_news_tech_!_想从事UI设计，该怎样入手呢？_!_
+6552467922525094152_!_109_!_news_tech_!_轴承安装方法有哪些？_!_
+6552338743234134279_!_109_!_news_tech_!_如果腾讯、阿里开始做农资了，你觉得会怎么做？_!_
+6552333336323293453_!_109_!_news_tech_!_马化腾说腾讯没有梦想，你还见过哪些牛人说过牛逼的话？_!_
+6552293583481733646_!_109_!_news_tech_!_中国移动5G联合创新中心天津开放实验室挂牌_!_物联网,中国移动,天津,实验室,NB-IoT
+6552140079937094152_!_109_!_news_tech_!_联通最便宜的全国不限流量套餐，今天真来了！_!_中国联通,冰淇淋,爱奇艺,百度,套餐
+6552242906009174285_!_109_!_news_tech_!_2.5平的电线可以承受多少电器？_!_
+6552307284280606980_!_109_!_news_tech_!_现代化牛场都要什么设备？_!_
+6552492929053622541_!_109_!_news_tech_!_什么叫做深度学习框架，其作用是什么？_!_
+6552330172597534990_!_109_!_news_tech_!_人类会不会进化到遗传记忆？_!_
+6552294537941745928_!_109_!_news_tech_!_红米note5待机一晚耗电10%正常吗？_!_
+6552493107273794052_!_109_!_news_tech_!_每天数字货币要闻（2018-5-6）_!_虚拟货币,数字货币,区块,比特币,区块链
+6552496764769796365_!_109_!_news_tech_!_如何评价文章《腾讯没有梦想》，腾讯是否真的在丧失产品能力变成投资公司？_!_
+6552368780192776462_!_109_!_news_tech_!_重新定义“智能汽车”，MARVEL X能成为荣威的“史册传奇”吗？_!_
+6552395199526469902_!_109_!_news_tech_!_联通10010人工服务老是打不通是什么原因？_!_
+6552316495769108744_!_109_!_news_tech_!_刺激战场既然这么火，为什么要投放机器人？_!_
+6552281863027687683_!_109_!_news_tech_!_为什么有人说前置指纹优于后置指纹？_!_
+6552354735230288391_!_109_!_news_tech_!_腾讯没有梦想引发轰动，马化腾刘炽平亲自回应！你赞同谁的说法_!_软银,腾讯没有梦想,英特尔,马化腾,刘炽平,腾讯
+6552127936906920200_!_110_!_news_military_!_西方制裁俄罗斯，俄罗斯靠什么撑了这么久？_!_
+6552446682146013699_!_110_!_news_military_!_二战维京师弹尽后用牙齿和拳脚死扛苏军_!_维京师,党卫队,骷髅师
+6552387691567120910_!_110_!_news_military_!_美媒：003型航母已开工，装国宝科学家颠覆性成果，改变游戏规则_!_辽宁舰,型航母,马伟明,航空母舰,核动力装置
+6551976643383001352_!_110_!_news_military_!_美国没有装备仿制的MG42机枪吗，为什么？_!_
+6552455787602510087_!_110_!_news_military_!_特朗普的淡漠反应，背后复杂而巨大的利益牵扯，美国会禁枪吗？_!_
+6552363519218548999_!_110_!_news_military_!_在战场上击倒敌人需要补枪吗？_!_
+6552451903429542414_!_110_!_news_military_!_全新战术大放异彩！俄神器促力，叙政府军猛烈攻击，反叛武装溃逃_!_反叛武装,叙利亚政府军,叙政府军,战术,巷战
+6552281748112146692_!_110_!_news_military_!_联合国安理会改革废除一票否决权，会导致什么后果？_!_
+6552469487411528195_!_110_!_news_military_!_对越反击战打草惊蛇战术妙，坑死越军1个加强排_!_掩蔽部,马旭旺,交通壕,手榴弹,越军
+6552268094293672200_!_110_!_news_military_!_伊拉克战争时，萨达姆为什么会将其强大的空军战机全部埋葬，弃而不用？_!_
+6552333209248465421_!_110_!_news_military_!_神盾的2000吨小舰，火力居然比中国054A还猛，我害怕_!_导弹,SH-60,巴拉克,以色列,护卫舰,直升机
+6552294441657303565_!_110_!_news_military_!_巴勒斯坦抗议者用弹弓击落2架以色列军用无人机_!_风筝,抗议者,巴勒斯坦,无人机,以色列
+6552380179719651592_!_110_!_news_military_!_AK47为什么能成为枪中之王？_!_
+6552260159534531076_!_110_!_news_military_!_烟火四起，俄罗斯家门口就迎来16国大军_!_爱沙尼亚,北约,莫斯科,俄军,演习
+6552410390326673928_!_110_!_news_military_!_中国2.0版航母战斗群浮出水面，“电飞鲨”放弃太行赶进度_!_航母编队,电飞鲨,电子战,歼-15,电子战机,导弹驱逐舰
+6552256178934514190_!_110_!_news_military_!_图看军事：美军C-130大力神，服役时间最长最优秀的运输机！_!_c-130大力神,C-130,美军,c-130运输机,运输机
+6552464647876772099_!_110_!_news_military_!_波兰捷克等国欢迎美国驻军吗，为什么？_!_
+6552453585018290445_!_110_!_news_military_!_如何评价满清八旗最后的名将——僧格林沁？_!_
+6552275841303183885_!_112_!_news_travel_!_2018河州牡丹文化月游客高峰再度来袭_!_芬芳馥郁,牡丹,临夏市,东公馆,五一小长假
+6552476694710583560_!_112_!_news_travel_!_世界第一伟人是谁？_!_
+6552302072409096708_!_112_!_news_travel_!_美国洛杉矶迪士尼乐园Disneyland一日游攻略_!_美国,洛杉矶迪士尼乐园,小小世界,米老鼠,迪士尼乐园
+6552451059296502286_!_112_!_news_travel_!_秋名山车神都虚的！重庆这些公路只有老司机才能驾驭，风景美如画_!_仙女山,周家山大桥,重庆,武陵山,大峡谷,失孤
+6552402893389955332_!_112_!_news_travel_!_洛阳当地人最认可的烩面店是哪些？_!_
+6552454672320299533_!_112_!_news_travel_!_杭州出发2h去这个初夏度假地，乐享田园星空轰趴、负氧离子爆表_!_淳安,负氧离子,淳安县,杨梅酒,千岛湖
+6552292758680240643_!_112_!_news_travel_!_几百米之上的玻璃栈道，让人眼发晕腿打颤，游客绝壁凌空囧态百出_!_长江三峡,天门山,洛阳白云山,玻璃栈道,东方IC
+6552379698930778376_!_112_!_news_travel_!_吉隆坡机场回中国的时候，不是在免税店买的物品可以托运吗？_!_
+6552464306166825475_!_112_!_news_travel_!_天上阿里，最精华的都在这个三国交界的边境县_!_普兰,玛旁雍措,普兰县,孔雀河,尼泊尔
+6552408048676110861_!_112_!_news_travel_!_「倡议书」关于广大市民推介临夏牡丹的倡议书_!_临夏市,临夏,牡丹文化,紫斑牡丹,河州牡丹
+6552457669427331588_!_112_!_news_travel_!_长沙免费景点大集合！不花一毛钱就能玩遍大长沙！_!_湿地公园,洋湖湿地,长沙,烈士公园,湘江
+6552388359367426573_!_112_!_news_travel_!_营口：“五一”期间 鲅鱼圈区接待游客近30万人次_!_五一,月亮湖,望儿山,游客量,鲅鱼圈区
+6552378788221551111_!_112_!_news_travel_!_山海万宁——大洲岛_!_海南,燕窝岛,四脚蛇,万宁,大洲岛
+6552359527159693837_!_112_!_news_travel_!_情报站｜泰安女游客一家半夜迷路，热心护林员护送下山_!_齐鲁壹点,检查站,爬泰山,乡土泰安,护林员
+6552461579391074830_!_112_!_news_travel_!_细看不似人间有_!_散文选刊,华夏散文,笔架山,牧野,杜鹃,星星,观景台,福建乡土,杜鹃山,井冈山,云海
+6552482020335812867_!_112_!_news_travel_!_南宁有哪些有特色的小酒馆？_!_
+6552128056708825352_!_112_!_news_travel_!_咸阳的袁家村到底是一个什么样的地方？_!_
+6552279048242856461_!_112_!_news_travel_!_全媒体记者（大咖）聚焦尧治河 见证中国最美乡村新发展_!_博物馆,尧治河村,风景区,尧治河旅游,旅游社
+6552382374351470851_!_113_!_news_world_!_如果美国打伊朗，最快可以多久解决战斗？_!_
+6552424284717842948_!_113_!_news_world_!_基拉韦厄火山喷发 民防部门：火山喷发没有减弱迹象_!_夏威夷,美国,火山喷发,基拉韦厄
+6552355473679450631_!_113_!_news_world_!_履行《板门店宣言》韩朝双方已完成扩音广播设备拆除工作_!_板门店宣言,韩朝,金正恩,文在寅,扩音广播
+6552440638007149064_!_113_!_news_world_!_中国和印度的经济差距怎么样，印度的邦跟中国县级市比怎么样？_!_GDP,武汉市,超过印度,印度,县级市
+6552372902853869832_!_113_!_news_world_!_为什么很多女孩喜欢嫁非洲黑人？_!_
+6552306959272378887_!_113_!_news_world_!_萨达姆究竟多有钱？看到这些照片后，网友都说难以置信_!_AK47步枪,伊拉克战争,伊拉克,萨达姆
+6552145153698562567_!_113_!_news_world_!_美国后院起火，反对派内部冲突 俄媒：美在叙势力只是一盘散沙！_!_逊尼派,叙利亚政府军,反对派,阿萨德,库尔德武装
+6552377966142161156_!_113_!_news_world_!_美国撤出北约的可能性有多大？_!_
+6552440593841127683_!_113_!_news_world_!_美国真的强大到可以在全球进行经济掠夺了吗？_!_
+6552281574451184141_!_113_!_news_world_!_各国如何与特朗普打交道？BBC：这些“奇招”了解一下_!_内塔尼亚胡,马克龙,默克尔,第一家庭,特朗普
+6552369456562045443_!_113_!_news_world_!_以色列竟强逼俄罗斯报恩！原来是为打伊朗扫清障碍_!_国防部长,伊朗,内塔尼亚胡,以色列,俄罗斯
+6552449313690091780_!_113_!_news_world_!_美国法院判伊朗赔付9·11遇难者家属逾60亿美元，你怎么看？_!_
+6552446626718286343_!_113_!_news_world_!_利比亚至今混战，卡扎菲众多女保镖去向成谜，结果超出预知_!_美国为首西方国家,利比亚战争,利比亚,卡扎菲,女保镖
+6552492608038371592_!_113_!_news_world_!_菲律宾众议长表示应废除联合国，建立亚洲联合国，有何深意？_!_
+6552322143760154888_!_113_!_news_world_!_哪位退休的教师领到百分之百退休金？_!_
+6552388881138843918_!_113_!_news_world_!_可以娶巴基斯坦的女士吗？_!_
+6552410428733915405_!_113_!_news_world_!_日本和以色列为什么会在贫瘠的土地上建立起高度发达的国家？_!_
+6552140891362951693_!_113_!_news_world_!_历史上的今天：5月6日_!_法国,柏林盖姆条约,捷克斯洛伐克,大亚湾核电站,邮票,排华法案,克莱斯勒公司
+6552344593516462595_!_115_!_news_agriculture_!_小麦减产已成定局？3天内已有大面积降水，农民需提前做好准备_!_倒春寒,小麦,扬花期,叶面肥,大兴安岭
+6552386299368571405_!_115_!_news_agriculture_!_很多农民都不愿生二胎，却不知独生子女即将面对这三大难题！_!_二胎政策,独生子女,独生家庭子女,生二胎,农村
+6552474270042161415_!_115_!_news_agriculture_!_农村出身的朱之文为什么回到农村不受大家欢迎？_!_
+6552329476510843399_!_115_!_news_agriculture_!_山坡上的房子_!_葡萄园,葡萄牙,山坡,橄榄树
+6552393441370702083_!_115_!_news_agriculture_!_有50万的资金买车还是买地？_!_
+6551586778266992904_!_115_!_news_agriculture_!_夏播玉米，种肥同播，用什么化肥好呢？_!_
+6552014847167431175_!_115_!_news_agriculture_!_即将进入插秧期，如何做好水稻移栽前准备！少花钱不得病！_!_秧苗,吡蚜酮,潜叶蝇,苗床,低温冷害
+6552488064533398029_!_115_!_news_agriculture_!_股市分析：从猪肉跌至10年低位看到的投机机会！_!_养猪,生命线,猪肉价格,温氏,猪肉出场价
+6552292095707578637_!_115_!_news_agriculture_!_龙虾田水深多少最好？_!_
+6552369250676245006_!_115_!_news_agriculture_!_芦笋种植年入百万！但这还不能满足他的野心……_!_芦笋,封丘县,新农民,王文利,合作社
+6552334640290464269_!_115_!_news_agriculture_!_农业农村部部长韩长赋莅临我校调研指导_!_韩长赋,生物农药,农业,南京农业大学,农村部
+6552350445929497102_!_115_!_news_agriculture_!_什么样的地方才适合诱蜂？有经验的蜂农选择诱蜂位置有什么特点？_!_蜂部落,诱蜂地点,蜜蜂
+6552354805610709517_!_115_!_news_agriculture_!_大叔野外捡到三只鸡蛋，本以为是意外收获，结果差点有牢狱之灾_!_鸡蛋,维护动物
+6552453415346110983_!_115_!_news_agriculture_!_农民2018年土地确权后，20岁以下没土地的人怎么办？_!_土地确权,在城镇,城镇化
+6552289690060325127_!_115_!_news_agriculture_!_雹灾对果树盆景的危害有哪些？_!_
+6552308371167379975_!_115_!_news_agriculture_!_“蒜你狠”变成“蒜你惨”，魔都蒜头最低5毛一个，不到最高时的一半_!_蒜头价格,蒜你狠,周到君,蒜头,邳县
+6552339791680111108_!_115_!_news_agriculture_!_农村非常普遍的野草，却有三大神奇功效，你肯定见过_!_李时珍,墨旱莲,中药材,城里人,农村
+6552408619315364099_!_115_!_news_agriculture_!_上联：猪不能没有毛，下联怎么对？_!_
+6552273228662833677_!_115_!_news_agriculture_!_日照市进一步优化完善农机购置补贴政策_!_产品认证,农机,购置补贴,农业生产经营组织,农业
+6552470222509441284_!_115_!_news_agriculture_!_农村小孩能不能到城里读书？_!_
+6552446670129332483_!_116_!_news_game_!_2000左右买什么手机？_!_
+6552400827846230541_!_116_!_news_game_!_王者荣耀：S11值得买的这几位13888英雄，要把钱花在刀刃上！_!_程咬金,出神入化
+6552411724840960526_!_116_!_news_game_!_王者荣耀：魔铠上分攻略，最暴躁的备战铭文、出装你敢用吗？_!_
+6552389379908698637_!_116_!_news_game_!_炉石传说：这套青玉萨在狂野环境胜率高达67%，你还不来试试？_!_萨满,2x,古尔丹,钢炼,飞刀杂耍者,青玉
+6552288709075534343_!_116_!_news_game_!_绝地求生：最想要的吃鸡神器_!_游戏,绝地求生
+6552359942437732867_!_116_!_news_game_!_王者荣耀哪些英雄大招技能最强呢？嬴政第二，第一神技能犹如外挂_!_嬴政,高渐离,宫本武藏,老夫子,姜子牙
+6552475783963607555_!_116_!_news_game_!_天龙八部手游：天命玄凤霸榜当之无愧！_!_战力,天命玄凤,九黎妖虎,战力排行榜,PVP
+6552281828340793860_!_116_!_news_game_!_英雄联盟移速最快四大打野，剑圣仅排第三，第一没人追得上_!_龙血武姬,打野,英雄联盟,无极剑圣,W技能
+6552338202777092366_!_116_!_news_game_!_qq飞车手游中为什么针尖那么不受人待见？_!_
+6552249094805389582_!_116_!_news_game_!_没有经验可以做什么店？_!_
+6552317693150626051_!_116_!_news_game_!_你支持迷你还是mc？_!_
+6552335946916823559_!_116_!_news_game_!_CFPL赛程，久离CF的你还会关注吗？_!_CF,直播间,龙珠TV,战队,CFPL
+6552140481499759112_!_116_!_news_game_!_传闻：《尼尔 机械军团》将登陆Xbox One平台_!_白金工作室,动作游戏,登陆Xbox One,Enix,Square,龙珠斗士Z,尼尔机械军团
+6552407247991865863_!_116_!_news_game_!_绝地求生刺激战场：跑毒载具选哪个？这些载具参数你必须懂_!_刺激战场,蹦蹦车,载具,摩托车,吉普车
+6552377993577103879_!_116_!_news_game_!_“天选之人”遭雷劈？你是否在装X，这个道具帮你检测……_!_阿拉德,圣骑士,神秘力量,勇士们,命运硬币
+6552295670261219847_!_116_!_news_game_!_阴阳师超鬼王即将结束 谁是最强特攻式神_!_特攻式神,SSR,青坊主,BOSS,大天狗
+6552308672913998344_!_116_!_news_game_!_《海王星》新作追加发售SWITCH版_!_海王星,Universe,Switch,RPG,勇者海王星,光之子
+6552266799612690701_!_116_!_news_game_!_做棋牌游戏代理真的靠谱吗？_!_
+6552360917512749571_!_101_!_news_culture_!_原来贾蔷是这么搞定贾珍贾蓉父子的_!_贾珍,秦钟,贾蔷,贾蓉,尤二姐
+6552339898785857796_!_101_!_news_culture_!_没文化、没技术，如何找出路？_!_
+6552335406006796558_!_101_!_news_culture_!_为什么好多人都说儒家思想祸害中国？_!_
+6552270980155179523_!_101_!_news_culture_!_历代名家书法精品——赵孟頫《灵隐大川济禅师塔铭》_!_柳公权,赵孟頫,书法,托古改制,上海博物馆
+6552396385910522125_!_101_!_news_culture_!_你愿意穿着汉服出门吗？_!_
+6552466308762435853_!_101_!_news_culture_!_上联：田在心上思故乡，求下联？_!_
+6552366439494320644_!_101_!_news_culture_!_莎士比亚名言精选（爱情篇）——以爱的名义靠近你_!_莎翁,鸡争鹅斗,桃花源,钥匙孔,丘比特
+6552346500964286734_!_101_!_news_culture_!_为啥一根破烂的金丝楠木能值2000万？_!_
+6552415207472234756_!_101_!_news_culture_!_为什么少林寺没有参加两次华山论剑？_!_
+6552242073288835341_!_101_!_news_culture_!_上联：杯中有酒赏明月，怎么对下联？_!_
+6552278731270914318_!_101_!_news_culture_!_上联：墙上画马不能骑，怎么对下联？_!_
+6552144104124317960_!_101_!_news_culture_!_上联：江中鱼东游西藏，下联怎么对？_!_
+6552352999526629640_!_101_!_news_culture_!_出联：山里人，水里人，山山水水人中人，如何接联？_!_
+6552313128724463886_!_101_!_news_culture_!_上联：苦雨虐荷，岸边老柳长垂泪；如何对下联？_!_
+6552309457248846343_!_101_!_news_culture_!_千古绝对！上联：“老牛吃嫩草”，下联太经典了_!_上联,我国历史,老牛吃嫩草,唐伯虎点秋香,下联
+6551307394989687048_!_101_!_news_culture_!_如何读懂一个上市公司发布的年报和财务报表，有哪些重要指标？_!_
+6552331485846700551_!_101_!_news_culture_!_欧阳询《皇甫诞碑》，可为初学者借鉴，甘苦之言_!_王世贞,皇甫诞碑,杨士奇,翁方纲,孙承泽,续书断,欧阳询,皇甫君碑
+6552367402674290957_!_101_!_news_culture_!_上联：彼岸花开处，下联怎么对？_!_
+6552379450711867918_!_101_!_news_culture_!_神仙住在张家界 历史留在云朝山_!_张三丰,金坪乡,云朝山寺,慈利县志,蜡烛山,廪贡生
+6552353335062561038_!_101_!_news_culture_!_上联：月下清贫花富贵，如何对下联？_!_
+6552359074594291975_!_102_!_news_entertainment_!_华晨宇是否会成为新一代歌手代表，成长为划时代人物？_!_
+6552360688969318920_!_102_!_news_entertainment_!_女明星产后患上“妈妈手”！这病或许你也有_!_熊黛林,腱鞘炎,双胞胎,陈嘉桦,郑云灿
+6552437130000335363_!_102_!_news_entertainment_!_景甜厉害了！恋爱之后体重过百了，网友：张继科真会疼人_!_萌萌哒,景甜,社交网站,张继科,丸子头
+6552363917828424206_!_102_!_news_entertainment_!_李嘉欣价值近6亿的豪宅曝光，网友：过的像电视剧_!_言情剧,巴厘岛,豪宅,李嘉欣,许晋亨
+6552449539633054215_!_102_!_news_entertainment_!_《风云1》第二梦到底是不是明月，5个完美的理由告诉你真正答案_!_风云1,夜行衣,聂风,武侠剧,风云,第二梦,明月
+6552249563921515015_!_102_!_news_entertainment_!_国内首档体育跨界综艺秀《高能卡位·世界波》点映礼在京举办_!_竞技体育,高能卡位·世界波,高晓松,东方IC,世界杯
+6552256312346935816_!_102_!_news_entertainment_!_王子文工作室发严正声明，王子文贴心安慰助理：委屈你了！_!_谁是我爸爸,女助理,严正声明,王子文,女演员,恐怖片
+6552401015948182019_!_102_!_news_entertainment_!_娱乐圈最珍贵的友谊，最后两个因为一个女人而反目成仇！_!_黄晓明,马天宇,赵薇,欢乐颂,房祖名
+6552311144831255047_!_102_!_news_entertainment_!_费启鸣抖音走红，邀请上快本后颜值遭质疑？原因是有幕后推手_!_快乐大本营,费启鸣,颜值,粉丝,戴眼镜
+6552430265094898184_!_102_!_news_entertainment_!_180506 网友造谣李易峰所乘车辆撞人 川影发布说明澄清_!_峰峰,赵半生,拔河比赛,动物世界,李易峰
+6552271353871860237_!_102_!_news_entertainment_!_学霸晒成绩 女神晒自拍 土豪晒有钱 模特晒身材 我晒太阳_!_
+6552407797923840519_!_102_!_news_entertainment_!_赚钱也没时间花，娱乐圈有多少像张钧甯这样的人？钱都去哪了？_!_粉丝,没时间,张钧甯,娱乐圈
+6552488783126725123_!_102_!_news_entertainment_!_娱乐类小说，走进这个世界最大的名利场_!_韩娱之勋,雨果-兰开斯特,名利场,综艺之谐星传奇,金酸莓奖,少女时代,好莱坞
+6552321863354155534_!_102_!_news_entertainment_!_昆凌和别的男人穿情侣装，周杰伦吃醋_!_昆凌,周杰伦,Anyone,情侣装
+6552290417499439374_!_102_!_news_entertainment_!_《绝地求生刺激战场》跑毒遭遇轰炸区是应该进房子，还是照常前进？_!_
+6552374396328083972_!_102_!_news_entertainment_!_明星拍戏避暑方法，鹿晗用风扇，赵丽颖吃冰棍，她却被骂耍大牌！_!_赵丽颖,戚薇,杨颖,冰棍儿,古力娜扎
+6552402748678078990_!_102_!_news_entertainment_!_光头好看的人是真的好看，那些剃了光头的明星，最后一个最帅！_!_吴亦凡,极限特工:终极回归,Men ' s Health,夏有乔木雅望天堂,西游伏妖篇,PO朝霆,谢霆锋,方谬神探,小鱼儿与花无缺,制作人,Statham,少年四大名捕,新楚留香
+6552465065163882759_!_102_!_news_entertainment_!_为什么录制《快乐大本营》的观众们没有爆料过？_!_
+6552383359840944648_!_102_!_news_entertainment_!_谭维维未婚夫陈亦飞发文报平安，还隔空喊话感谢老婆真幸福！_!_陈亦飞,谭维维,病号服,演唱会,冈仁波齐
+6552316051219022349_!_103_!_news_sports_!_温格主场能否完美谢幕？来看看博彩公司的意图！_!_温格,伯恩利,勒沃库森,欧联杯,阿森纳
+6552339438062535181_!_103_!_news_sports_!_德罗赞取尴尬数据三节“打卡” 他离“超级巨星”还差得有点远_!_德罗赞,超级巨星,詹姆斯,正负值,半决赛
+6552368261323817476_!_103_!_news_sports_!_这才是老詹一见猛龙就往死里打的真相，真是让人拔凉拔凉的！_!_猛龙,詹姆斯,德罗赞,老詹,凯西
+6552486328154456580_!_103_!_news_sports_!_无兄弟不篮球，胜利并不是所有，身边有一个好兄弟陪你分担，足矣_!_凯尔特人,猛龙,德罗赞,詹姆斯,骑士
+6552380762564329731_!_103_!_news_sports_!_你最喜欢的NBA巨星是谁，讨厌的巨星又是谁，为什么？_!_
+6552288557321421320_!_103_!_news_sports_!_球队开心果！火箭官方晒周琦为塔克唱生日歌视频：真奇妙_!_周琦,社交网站,塔克,生日歌,火箭队
+6552367963243020548_!_103_!_news_sports_!_猛龙VS骑士G3，詹姆斯绝杀的那一球，骑士队为什么不选择发前场球？_!_
+6552389139994509838_!_103_!_news_sports_!_珍藏版！2018俄罗斯世界杯赛程表，世界杯32强巡礼！_!_卡瓦尼,俄罗斯世界杯,卢卡库,世界杯,欧洲杯
+6552477742829404686_!_103_!_news_sports_!_英冠第二支冲超队：百年名队重返英超，刘德华舅舅被骂4年终得正名_!_卡迪夫城,英超,陈志远,刘德华,英冠
+6552401961856008711_!_103_!_news_sports_!_继詹皇、King James、老北京之外，詹姆斯又得一个新绰号_!_季后赛,詹姆斯,绰号,LBJ,勒布朗
+6552825989229969928_!_103_!_news_sports_!_官方：里贝里与拜仁续约一年_!_里贝里,拜仁
+6552461874707824900_!_103_!_news_sports_!_奥兰多，安迪尔森，詹姆斯，穆谢奎，谁对大连足球的贡献更大？_!_
+6552284438108045576_!_103_!_news_sports_!_科尔不满隆多垫脚库里欲向联盟投诉隆多，要是莱昂纳德听到这消息会有什么感觉呢？_!_
+6552253744573055502_!_103_!_news_sports_!_恒大到底输在哪里？！0:3之后卡帅一番话给出输球真因！_!_卡纳瓦罗,高拉特,广州恒大,郑智,思想意识
+6552352784195256846_!_103_!_news_sports_!_他是中超唯一一个巴西主力 或成世界杯赛场中超最亮眼的名片_!_中超,巴西国家队,胡尔克,奥古斯托,巴西队
+6552267324349481223_!_103_!_news_sports_!_世乒赛女团决赛，世界第一快刘诗雯2:3爆冷输球。你怎么看？_!_
+6552411197709222407_!_103_!_news_sports_!_定个小目标？巴萨盼保不败+挫皇马锐气 助梅西争欧洲金靴_!_国家德比,诺坎普,巴萨,特尔施特根,皇马
+6552386595176055043_!_103_!_news_sports_!_老将内内的出场时间骤然增加不少，而安德森跌出轮换，德帅此意为何？_!_
+6552400440418370055_!_103_!_news_sports_!_中国收入最高的球星：三度结婚终娶小17岁女友，被誉为足球领袖_!_足球,最佳球员,范志毅,李茏怡,张梦瑾
+6542749623163814152_!_104_!_news_finance_!_6月30号后取消第三方支付直接扣账，支付宝还能直接用于支付吗？_!_
+6550262917411897869_!_104_!_news_finance_!_存100万赚利息，就可以不上班了吗？想得太美_!_余额宝,通货膨胀
+6538151602212045316_!_104_!_news_finance_!_一个故事看懂金融危机时钱都去哪里了？_!_金融危机,灰天鹅,零和游戏,社会财富,杯子
+6542255775551586820_!_104_!_news_finance_!_如果把500万人民币存进银行，光靠利息能过上什么样的日子？_!_财付通,银行,利息收入,支付宝,余额宝
+6516386854873350414_!_104_!_news_finance_!_如何判断一个人是聪明还是笨？_!_判断,马云,创业者
+6549481050207158787_!_104_!_news_finance_!_十五万存银行 每年多少钱 算一下！_!_国有银行,支付宝,银行存款,银行
+6551523327175819783_!_104_!_news_finance_!_行长很着急：浦发银行成唯一利润负增长银行，造假竟造出零不良率_!_信用卡,兴业银行,银行业,浦发银行,营改增,民生银行,招行
+6551537737671901699_!_104_!_news_finance_!_房企和炒房客“复盘无望”降价抛售将加快“房价下跌”_!_房地产,关系户,任志强,四万亿,经济学家
+6517112366839103758_!_104_!_news_finance_!_社保不想续交了，已经交的钱能不能退？_!_养老金,社保
+6550791183059124749_!_104_!_news_finance_!_「早参」姚振华胜利大逃亡_!_格力空调,格力电器,王晓峰,机器人,姚振华,董明珠
+6550622577633526275_!_104_!_news_finance_!_揭密1990年日本大股灾房、地产泡沫真面目_!_股市,货币战争,日元,股票市场,泡沫经济,日经平均指数,日本经济新闻,日经指数,广场协议,股市泡沫,金融战败,日本股市,可口可乐公司,日本经济
+6550970745827099149_!_104_!_news_finance_!_骗子不够多，傻子永远都够用_!_不劳而获,风险投资
+6494848473853067534_!_104_!_news_finance_!_为什么很多在工厂上班的人，甘愿每月拿着几千元的固定工资？_!_外面闯荡,工厂上班,人生苦短
+6551970932611613198_!_104_!_news_finance_!_如果美国将国债一次性还清，会发生什么事！_!_美国政府,国债,美国国债,美国
+6552424421334712579_!_104_!_news_finance_!_银行卡没有钱，如果一直不用，每年都扣年费等等费用，十年二十年后，银行会主动来要费用吗？_!_
+6552432304445194499_!_106_!_news_house_!_一套150平二手房，售价86万，过户税费额外需交多少钱？_!_
+6552429910357442823_!_106_!_news_house_!_购买二手房，卖家反悔？_!_
+6552327546480886279_!_106_!_news_house_!_芜湖再出人才新政 购房最高可补贴200万元_!_安家费,芜湖市,购房,人才引进,战略性新兴产业
+6552319496407744782_!_106_!_news_house_!_新手买房，是找中介好还是自己去售楼部买更好？有什么需要注意的？_!_
+6552368278197502211_!_106_!_news_house_!_盐城市大洋新村会拆迁吗？_!_
+6552434903974150414_!_106_!_news_house_!_签署《离岗通知单》约定商业秘密保密义务是否有效？_!_
+6552344361282044164_!_106_!_news_house_!_房价走势如何？_!_
+6552493722775323143_!_106_!_news_house_!_交房租还是还网贷？“黑中介”捆绑网贷平台设“套”_!_房租,租房,元宝e家,惠人贷,37号,房地产经纪,中介公司,房产中介,租户,昊园恒业
+6552390300902359310_!_106_!_news_house_!_合肥的房价2018年还会涨多少？_!_
+6552349946765377796_!_106_!_news_house_!_兄弟两人哥哥去世了，父母的三套房产给哥哥的儿子一套，嫂子还能去争夺别的家产吗？_!_
+6552458702098530567_!_106_!_news_house_!_山东日照2018年买了房子之后，房价跌了怎么办？_!_
+6552274670144455172_!_106_!_news_house_!_原聚场——东原线下社群实体空间在东原印未来盛大启幕_!_社群生活,东原,漫画家,原·聚,小剧场
+6552275773057663492_!_106_!_news_house_!_4月份济南这些热门楼盘拿到预售证！获批项目房价一览_!_中海,济南,城市之光,历城区,御景台
+6552387075763601671_!_106_!_news_house_!_四季如春的昆明有哪些冷知识？_!_
+6552249829626479118_!_106_!_news_house_!_房价涨那么高，为什么还有人抢着买？_!_房价,购房,楼市,买房
+6552281815418143239_!_106_!_news_house_!_为什么一夜之间冒出那么多全款买房的有钱人？_!_公积金,买房,楼盘,房源,房价
+6552288721285153037_!_106_!_news_house_!_巴中这个城市怎么样？_!_
+6552385981167698180_!_106_!_news_house_!_马云又出惊人言论，8年后房子最不值钱，他的话可信吗？_!_
+6552295225383977485_!_106_!_news_house_!_买美国哪种房子最赚钱？楼房别墅大比拼！｜小李抛砖_!_拥有产权,房产,房产税,购买者,别墅
+6552364298822222350_!_106_!_news_house_!_房价只涨不跌是正常？专家预测：20年后房价达到80万每平_!_房价,房地产,楼市,湘邮科技,城市房价
+6552267747584115213_!_107_!_news_car_!_奔腾T77这款概念车怎么样？_!_一汽奔腾,刹车灯,T77,北京车展
+6552474788969841159_!_107_!_news_car_!_帕纳梅拉！动感又不失优雅！_!_梅拉,帕纳,轿跑车,保时捷帕纳梅拉,保时捷
+6552336900458283534_!_107_!_news_car_!_聚焦未来｜专家热议汽车产业高质量发展_!_电动汽车,自主创新,汽车公司,汽车产业,中国汽车
+6552439569000694275_!_107_!_news_car_!_行车记录仪该淘汰了！如今义务造“8D魔术镜”大火，性能忒强劲_!_     ,收纳袋,防滑垫,坚固牢靠,抬头显示器
+6552409004218253575_!_107_!_news_car_!_2015版的长安奔奔还值得购入吗？_!_
+6552411851580244494_!_107_!_news_car_!_8万搞定，这四款SUV懂车的人都会买，让吹飞度的人脸红_!_瑞虎,宝骏,T1X,飞度,SUV,指导价,风光580,吉利远景
+6552341906464964868_!_107_!_news_car_!_改装汽车排气管，会增加动力吗？_!_
+6552482415632187661_!_107_!_news_car_!_老婆月薪4000多点，我8000左右，两个孩子幼儿园，房子没有压力，想买台迈腾，能不能养得起？_!_
+6552321599083643395_!_107_!_news_car_!_它是来自北京车展的一匹黑马，起售价八万！_!_北京车展,MPV
+6552468902658441479_!_107_!_news_car_!_机油滤芯有一次性和可换纸滤芯的，为什么有设计区别，哪种更好？_!_
+6552345871894184455_!_107_!_news_car_!_贴心小老公给爱车换上这几款车品，跑长途开夜车，样样有大用处_!_汽车蜡,密封条,降温剂,汽车漆,吸油器
+6552408852850016771_!_107_!_news_car_!_长安CS75这款车到底怎么样？看看车主400公里用车真实反馈吧！_!_软内饰,SUV,后备箱,长安,CS75,自动挡,长安汽车
+6552281753015288333_!_107_!_news_car_!_经常开车，车漆越开划痕越多！教你一招，1分钟消除划痕还不留印_!_抛光蜡,划痕修复,太阳纹,汽车划痕修复,划痕
+6552358749783196167_!_107_!_news_car_!_这货不到10万，很适合五一出游，空间也满足要求，值得入手！_!_GS4,传祺gs4,传祺GS4
+6552313574197297668_!_107_!_news_car_!_5月，河南限行城市及限行区域又有新变化，车主们速看！_!_河南,限行,机动车,机动车号牌,阿拉伯数字
+6552378342933266696_!_107_!_news_car_!_名爵ZS这款车使用感觉怎么样，为什么买这款车？_!_
+6552268217941754372_!_107_!_news_car_!_一脚油门奔驰变废铁 车子从20余米高山坡滚下_!_浙江,玉苍山,山坡,苍南县,车子
+6552479941508727044_!_107_!_news_car_!_大众朗行1.6自动挡用5w-30全合成机油还是5w-40半合成机油？_!_
+6552470666208084231_!_107_!_news_car_!_汽车发动机积碳问题真的不可避免吗？_!_
+6551221326260994568_!_107_!_news_car_!_几款适合年轻人的车，八万左右，月供1千就能行_!_福特福睿斯,大众捷达,福睿斯,比亚迪,此时此刻
+6549006548050379267_!_108_!_news_edu_!_孩子有这些情况，可能是脊柱侧弯！快看看你小孩中招没_!_肩胛骨,脊柱侧弯,矫形器
+6552353959267271171_!_108_!_news_edu_!_话题作文怎么写？送给你12个话题素材，用好了作文都会出彩！_!_命运交响曲,迷失方向,小草,我学会,一泓清泉,打开心,时间好比,象牙塔,奋斗者,出师表,买不到,心灵看世界
+6552483247467528456_!_108_!_news_edu_!_苏州大学全国排名第13位不太可能吗？_!_
+6552255481702777351_!_108_!_news_edu_!_在农村读书的要注意了！_!_绩效工资,寄宿制学校,农村,乡村教师
+6552434679067181315_!_108_!_news_edu_!_高考专业相同或者相近，院校应如何选择？_!_
+6552394018813116941_!_108_!_news_edu_!_人工智能有多火？全国首个人工智能课程已经中小学落地了！_!_人工智能,教育信息化2.0行动计划,新一代人工智能发展规划,朝阳区,初识语音识别,人工智能实践教程,人工智能科普教育,创造栗,编程
+6552295254597304589_!_108_!_news_edu_!_七年级男孩子学习不自觉怎么办？_!_
+6552386615522623758_!_108_!_news_edu_!_准研究生从现在到开学这段时间做些什么来赚点钱呢？_!_
+6552423056071983620_!_108_!_news_edu_!_北师大版一年级数学期末综合复习题，非常实用_!_年级,期末考试,北师大,复习题
+6552299727528919555_!_108_!_news_edu_!_“小豆包”进幼儿园科普消防知识_!_幼儿园,消防知识,史家小学,小豆包,妹妹们
+6552223855392325895_!_108_!_news_edu_!_如何看待大学学习问题上的博广与精深？_!_
+6552412631821451779_!_108_!_news_edu_!_何诚总代表与图兰大学进行联合办学项目会谈_!_陕西,代表处,总代表,阿拉木图,拉赫曼
+6552388200923398670_!_108_!_news_edu_!_「学族教育」课辅品牌那么多，在线也有不少，到底哪家合适？_!_在线教育,在线学习,教育软件,多样性,大数据
+6552294223763210510_!_108_!_news_edu_!_哪种软件适合成人学习英语？_!_
+6552421154059976963_!_108_!_news_edu_!_马上就高考了，怎么复习？_!_
+6552356909532316174_!_108_!_news_edu_!_浙江理工大学_!_二级学科,实验班,应用心理学,一级学科,硕士学位
+6552339089331323400_!_108_!_news_edu_!_怎样打开正确的教育方式？_!_翻转式学习,教学方式,教与学,体育课,重复式
+6552280887839424772_!_108_!_news_edu_!_如今社会如何看待专科、本科、研究生学历？_!_
+6552232462766833924_!_108_!_news_edu_!_你怎么看待父亲精神分裂，母亲下肢残疾的他高考却被清华录取？_!_
+6420884696409833729_!_109_!_news_tech_!_低调上架，小米新机4G+64G售1099，不用抢你会不会买？_!_小米
+6538212901226807811_!_109_!_news_tech_!_华为手机越来越卡？赶紧开启这2个开关，用1辈子也不会卡！_!_华为手机,安卓手机
+6534171660218532103_!_109_!_news_tech_!_性别仿真机器人出现，我们还需要异性吗？_!_
+6550649606743523854_!_109_!_news_tech_!_章泽天告诉我们：要想嫁得好，首先你得好！_!_金陵十三钗,嫁入豪门,刘强东,清华大学,奶茶妹妹,章泽天
+6553077243659158019_!_109_!_news_tech_!_iPhone X Plus真机现身：三摄对标华为 售价超一万 苹果真怕了？_!_iPhone X Plus,iPhone X,华为,iPhone 8,iPhone 4
+6532988689973248526_!_109_!_news_tech_!_华为手机音量小，一定要按这个按钮！赶快试试吧_!_音量,华为手机
+6551320109758022157_!_109_!_news_tech_!_网易——马化腾也打不死的小强，扫地僧般的存在_!_市场份额,丁磊,网易云音乐,腾讯,网易公司,亮剑
+6520759225692455171_!_109_!_news_tech_!_去非洲工作一年的人回来会有什么变化，年薪十多万值得吗？_!_智能手机,肯尼亚,在非洲,信用卡,乌干达
+6552788432274326020_!_109_!_news_tech_!_罗永浩13点亲自确认，5月15日，他要发布2款革命性产品_!_华为,小米,相声,罗永浩,极客公园
+6533060331210539524_!_109_!_news_tech_!_魅族手机这个功能就值300块，刚买手机就要学会，否则手机白买了_!_魅族手机
+6553055391956075021_!_109_!_news_tech_!_日本向中国提供半导体技术，是帮助中国对抗“科技战”还是其他？_!_智能手机,ARM控股,本土化,CPU,半导体
+6514550438421856516_!_109_!_news_tech_!_经常更换微信头像的人是怎样的一种心理？_!_微信,换个头,社交软件
+6552794616968512008_!_109_!_news_tech_!_FF 91抵达北京，为国产做准备，贾跃亭要归来了？_!_NBD汽车,乐视网,莫干山,贾跃亭,乐视汽车
+6553086894677164548_!_109_!_news_tech_!_华为手机关掉这几个开关，缩减手机耗电量，提高待机时间_!_待机时间,耗电量,华为手机
+6510467258563690759_!_109_!_news_tech_!_iphone6plus还能用几年？_!_iOS8,iPhone4,6Plus,诺基亚,iPhone
+6552296671647105283_!_110_!_news_military_!_美国动不动就拿撤军来威胁韩国、日本，日本、韩国为什么那么怕美国从他们那里撤军？_!_
+6552257741140787463_!_110_!_news_military_!_为什么打仗时，千万不能打跳伞的飞行员？_!_
+6552398831303000328_!_110_!_news_military_!_特朗普的淡漠反应，背后复杂而巨大的利益牵扯，美国会禁枪吗？_!_
+6552399990327607560_!_110_!_news_military_!_叙利亚政府军从一处叛军据点缴获的武器里面有俄制的装备，你怎么看？_!_
+6551964680766620167_!_110_!_news_military_!_一次可毁灭200座城市，全世界仅有一艘，中国想要但被拒绝_!_战略核潜艇,弹道导弹核潜艇,核弹头,俄罗斯,导弹
+6552436918041182734_!_110_!_news_military_!_西安无人机表演乱码 官方道歉称有定向干扰_!_编队飞行,应用技术,1374架,吉尼斯世界纪录,红五月西安城墙国际文化节,无人机编队表演
+6552357659255767299_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552149657508643075_!_110_!_news_military_!_叙利亚总统阿萨德会成为第二个卡扎菲吗？_!_
+6552267889171235080_!_110_!_news_military_!_女乒今天排兵布阵不合理，丁宁昨天刚打硬仗，今天应该打第一单打，你认为呢？_!_
+6552459066248003844_!_110_!_news_military_!_当今谁是世界上强国？_!_
+6552337874388254983_!_110_!_news_military_!_你怎么看待以色列总理出席俄罗斯5月9日的阅兵并会晤普京？_!_
+6552419885417234691_!_110_!_news_military_!_二战，德国如果在击败法国后，不发动巴巴罗萨战役与苏联开战，改为修养生息。与英国僵持能否改变战争走向？_!_
+6552241783231742215_!_110_!_news_military_!_女乒今天排兵布阵不合理，丁宁昨天刚打硬仗，今天应该打第一单打，你认为呢？_!_
+6552301334014460164_!_110_!_news_military_!_中国当今有哪些超级工程？_!_
+6552446083790799107_!_110_!_news_military_!_美国为何要打叙利亚？_!_
+6552285977530859783_!_110_!_news_military_!_联合国安理会改革废除一票否决权，会导致什么后果？_!_
+6550590566764118279_!_110_!_news_military_!_为了应对德军的狼群战术，盟军都采用了哪些手段？_!_
+6552260403437502979_!_110_!_news_military_!_俄罗斯要找中国帮忙？和辽宁舰有关，你觉得该同意吗？_!_舰载机飞行员,辽宁舰,戴高乐号航母,库兹涅佐夫号,俄罗斯
+6552276431555002638_!_110_!_news_military_!_美国为什么怕俄罗斯？_!_
+6552428144953917710_!_110_!_news_military_!_白杨导弹能否打死恐龙？_!_
+6552371185538040071_!_112_!_news_travel_!_护照有效期不足6个月能去普吉岛吗？_!_
+6552467463806648835_!_112_!_news_travel_!_刷完抖音完好想好想去南宁！_!_南宁,邕江,汽锅鸡,南宁市,中山路
+6552310562003353860_!_112_!_news_travel_!_你去纳米比亚花了多少钱？15天左右，人均2W够吗？_!_
+6552271262930960648_!_112_!_news_travel_!_大理的喜洲古镇有什么好吃好玩的？_!_
+6552437666573451783_!_112_!_news_travel_!_中国最热的地方—新疆吐鲁番火焰山_!_万佛山,火焰山,吐鲁番,千佛洞,馒头山
+6552479556672946691_!_112_!_news_travel_!_中国历史文化名镇、广西四大古镇之一大圩古镇走访记_!_公共厕所,大圩古镇,桂林市,灵川县,古镇
+6552444746864787975_!_112_!_news_travel_!_探讨：乡村旅游的N+1种新玩法，值得借鉴！_!_永春县,童玩节,展销会,乡村旅游,观光
+6552274779796144398_!_112_!_news_travel_!_书上说，地球上大量的雨水蓄满了火山口，形成海洋。雨水是淡水，海水为什么是咸的？_!_
+6552448004203217422_!_112_!_news_travel_!_中国最美大学排行榜，这两所学校远超北京大学_!_武汉大学,厦门大学,未名湖,北京大学,博雅塔
+6552422439437992461_!_112_!_news_travel_!_穷游一定要避开这个国家，出行打不到出租车，不预定连宾馆都没有_!_房价,摩纳哥,出租车,上流社会
+6552243465504162062_!_112_!_news_travel_!_永城的“紫薇大道”、“樱花大道”、“月季大道”您知道在哪里吗？_!_
+6552407468066996743_!_112_!_news_travel_!_趁年轻，去美丽的上川岛看一看_!_方济各,上川岛,圣方济各·沙勿略,看上川岛,张保仔
+6552479179651154184_!_112_!_news_travel_!_世界第一伟人是谁？_!_
+6552449092616716807_!_112_!_news_travel_!_刺激战场：连接两个最偏僻点的航线，知道它的并不多_!_军事基地,资源量,航线,别墅区,★★推荐
+6552363273616884238_!_112_!_news_travel_!_「评论」推进休闲农业与乡村旅游升级！致富多一条路！_!_乡村旅游,两者之间,农业生产,休闲农业,旅游业
+6552350547444236804_!_112_!_news_travel_!_旅顺二〇三樱花园缤纷花季_!_樱花树,旅顺,浪漫樱花,樱花园,樱花节
+6552306341015191821_!_112_!_news_travel_!_日本茨城机场附近有什么能玩的地方么？_!_
+6552353617058201863_!_112_!_news_travel_!_去越南旅行，什么时候去比较合适？_!_
+6552289923498508804_!_112_!_news_travel_!_秀丽多姿的山川溪水——浦江茜溪，金华人夏天最爱的旅游风景区_!_马头峰,中国书画,板凳龙,杭金衢,茜溪
+6552365901788742157_!_112_!_news_travel_!_中国第一水乡，周庄_!_周庄镇,贞丰,阿婆茶,周庄,沈万三
+6552267278623179277_!_113_!_news_world_!_奇幻！大叔生日当天退休买彩又中1000万_!_加拿大,大叔,温哥华
+6552236542352425223_!_113_!_news_world_!_如果美国总统特朗普辞职，会对整个世界带来什么影响？_!_
+6552246621692756487_!_113_!_news_world_!_中国和波黑互免签证协定将于5月底生效_!_签证,中华人民共和国政府,大使馆,波黑,普通护照
+6552132794091307277_!_113_!_news_world_!_怎么看因为性侵丑闻，二战后诺贝尔文学奖首次停摆？_!_
+6552473145339543816_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552432865538212104_!_113_!_news_world_!_犹太人为何一定要回到耶路撒冷地区建国？_!_
+6552136617774022919_!_113_!_news_world_!_如果特朗普遭弹劾下台，谁来接替当总统？_!_
+6552408811989107207_!_113_!_news_world_!_尚未成年的夏洛特小公主竟然主动亲吻别的男人_!_夏洛特,小王子,剑桥公爵,肯辛顿宫,Louis
+6552357731758506253_!_113_!_news_world_!_蒙古国人口为什么那么少？_!_
+6552459502833107208_!_113_!_news_world_!_印度这样放纵人口发展，最后明显是一大祸事，国家和政府为什么不加以控制？_!_
+6552289101763051780_!_113_!_news_world_!_像委内瑞拉这样的国家和人民，是怎么混成连饭都吃不起的，根本原因是什么，如何挽救？_!_
+6552321548299010308_!_113_!_news_world_!_有人认为当美国总统很辛苦，那为什么一些人还争着当？_!_
+6552321150502830605_!_113_!_news_world_!_揭开日本艺伎的神秘面纱，渐行渐远的背影风光不再_!_日本,艺伎,东方IC,表演艺术
+6552288101371216135_!_113_!_news_world_!_为什么泰国没有假药？_!_
+6552338433761608199_!_113_!_news_world_!_柬埔寨东北部发生中毒事件 已致6人死亡52人住院_!_桔井省,东北部,柬埔寨,中毒
+6552252459077599758_!_113_!_news_world_!_全年法定带薪节假日达28天！世界上假期最多的国家竟是它_!_红色高棉,金边,柬埔寨,柬埔寨人,法定节假日
+6552370562100887821_!_115_!_news_agriculture_!_黄土、黑土、红土哪一种土更适合农业发展？各自适合种植什么东西？_!_
+6552447121126064392_!_115_!_news_agriculture_!_小时候，在农村读书的时候，你做过最后悔的事情是什么？_!_
+6552268741919375880_!_115_!_news_agriculture_!_今年是个丰收年，世界上最大的水果，高速路上晒一晒！_!_果实,菠萝蜜,蛋白质,热带水果,丰收年
+6552307714851078669_!_115_!_news_agriculture_!_市场分析｜原药有涨有跌，2018年农药行情扑朔迷离？_!_原药,吡唑醚菌酯,价格下调,TC,氰氟草酯
+6552395344578085390_!_115_!_news_agriculture_!_辽宁省突破秸秆生物炭规模化应用关键技术 生产成本仅为国外同类产品的15%_!_化学肥料,秸秆炭化,辽宁省,规模化,土壤改良剂
+6552377171896173059_!_115_!_news_agriculture_!_这种“野味”农村很常见，扔一个，长一片，现在卖到15元一斤_!_农村,洋火姜,家家户户,植物,蔬菜
+6552294065751195907_!_115_!_news_agriculture_!_种植花生为什么要进行清棵，清棵的好处有哪些？_!_
+6552389455863349773_!_115_!_news_agriculture_!_这个夏季十分常见的野菜，吃起来像苋菜一样滑滑的，略有些酸！_!_五行草,房前屋后,马齿菜,辣椒油,马齿苋
+6552354652891906567_!_115_!_news_agriculture_!_农民卖山里产的稀罕物山货，品种繁多都是好东西，你喜欢食用它吗_!_
+6516023807784780302_!_115_!_news_agriculture_!_竹林套种大球盖菇优质高效栽培技术_!_栽培,盖菇,播种,农业废弃物,培养料
+6552386226983272963_!_115_!_news_agriculture_!_吃百香果，喝百香果汁不时髦啦！2018流行吃“香蕉百香果”_!_树葡萄,香蕉百香果,西番莲,嘉宝果,山竹
+6552478953825632525_!_115_!_news_agriculture_!_今年这轮“猪周期”，大跌之后必有大涨吗？_!_
+6552344072088977934_!_115_!_news_agriculture_!_巷道年久失修，严重影响出行_!_天然气,巷道
+6540598391699669511_!_115_!_news_agriculture_!_山东滕州 南沙河镇_!_食用菌,滕州市,山东,南沙河镇,山东省
+6552313628245099016_!_115_!_news_agriculture_!_中国农科院集成创新推出蔬菜绿色生产“4H”模式_!_我国蔬菜,设施蔬菜,中国农业科学院,集成模式,种植业
+6552368117018788365_!_115_!_news_agriculture_!_未来，这几种菇农，将成为困难户！_!_困难户,马马虎虎,菇农
+6552302443785355780_!_115_!_news_agriculture_!_农村在这个季节家家户户种的豌豆已经成熟了 那你们家的呢_!_豌豆,农村
+6552255917746815236_!_116_!_news_game_!_怎么看微博大V爆料《王者荣耀》掉帧卡顿等问题，是因为手机品牌保护费交的不够？_!_
+6552451155090211079_!_116_!_news_game_!_有什么唯美剧情的手机单机游戏？_!_
+6552431507284164877_!_116_!_news_game_!_截止至2018春季赛季后赛，谁是lck第一adc？_!_
+6552258906616234253_!_116_!_news_game_!_王者荣耀有哪些辅助可以出全输出装？_!_
+6552472916749976067_!_116_!_news_game_!_巴西赛区KBM战队：感谢粉丝和所有的一切_!_战队,MSI,SUP,巴西
+6552454191061664263_!_116_!_news_game_!_小心！小程序正在毁掉你的微信体验_!_游戏,小游戏,分享复活,手机qq
+6552403206482166024_!_116_!_news_game_!_如果可以，让你从DNF的NPC中选择一个带来现实，你会选谁？_!_
+6552328367541060104_!_116_!_news_game_!_DNF：安徒恩比卢克好在哪？为什么许多勇士都喜欢乌龟讨厌卢克_!_DNF,三神器,安图恩,碰运气,卢克
+6552417388816498951_!_116_!_news_game_!_lol现在哪个英雄单挑最厉害？_!_
+6552153411024847374_!_116_!_news_game_!_游戏圈的C位竟被这个男人占了 郭炜炜没想到自己“被出道”_!_何曾惧,VK,郭炜炜,剑网3,爱豆,制作人
+6552461304659968525_!_116_!_news_game_!_MSI泰国战队再现“黑科技” 主教练竟带“武林秘籍”登场_!_Pawn,H2K,ASC战队,化骨绵掌,泰国
+6552308929945141511_!_116_!_news_game_!_王者荣耀里钟馗的钩子和百里玄策的钩子哪个更可怕，你更怕哪个？_!_
+6552493737669296653_!_116_!_news_game_!_dnf玩家穿超大陆刷安徒恩屡屡被踢，朋友提醒后恍然大悟_!_地下城与勇士,dnf,地下城,游戏,超大陆
+6552344690736234765_!_116_!_news_game_!_到2018年5月为止，有哪些游戏被新闻正面报道过？_!_
+6552454194027036942_!_116_!_news_game_!_手残可以做游戏实况视频吗？_!_
+6552187546804158734_!_116_!_news_game_!_为什么暴雪不把dota做成手游？_!_
+6552381124373381636_!_116_!_news_game_!_官人某牙事件再度升级！官人：欠我的钱不要了，高管敢回应我吗？_!_LOL,官人
+6552361259767955715_!_116_!_news_game_!_lol中，用武器打不死盖伦该怎么办？_!_
+6552127043335618819_!_116_!_news_game_!_5月5日绝地求生天命杯4AM战队成绩不理想，你怎么看？_!_
+6552117871885943300_!_101_!_news_culture_!_马克思的宣传画 浓浓的历史感_!_上海人民美术出版社,人民美术出版社,马克思,巴黎公社,马克思主义
+6552299734453715204_!_101_!_news_culture_!_上联：江中鱼东游西藏，下联怎么对？_!_
+6552295618797109517_!_101_!_news_culture_!_上联：多心未必是好事；下联？_!_
+6552317589576483076_!_101_!_news_culture_!_开弓没有回头箭，下句怎么接？_!_
+6552406846357897735_!_101_!_news_culture_!_冷门好碑帖06：谢安行书手札，东晋超级男神的书迹_!_书品,谢帖赞,谢安,王羲之,书法,王献之,兰亭诗,论书,述书赋,书后品,书断,王僧虔,书谱
+6552401998052852232_!_101_!_news_culture_!_此人一眼看穿孙悟空的筋斗云，连太上老君都忌惮三分，不是菩提_!_太上老君,菩提祖师,筋斗云,孙悟空,如来
+6552127852232311054_!_101_!_news_culture_!_上联：层层叠叠多多彩，如何对下联？_!_
+6552350781935190275_!_101_!_news_culture_!_上句：但愿人长久，下句怎么对？_!_
+6552496396841255175_!_101_!_news_culture_!_上联：叹人生苦短，今宵难眠。下联怎么对？_!_
+6552133432544067848_!_101_!_news_culture_!_上联: 芦苇揺曳随风摆，下联如何对？_!_
+6552263412951286023_!_101_!_news_culture_!_玉笛吹来美人风，如何赐句？_!_
+6552303349872460036_!_101_!_news_culture_!_上联：沉鱼落雁千君醉，如何对下联？_!_
+6552472148961657096_!_101_!_news_culture_!_上联，朱雀桥边照飞燕，如何对下联？_!_
+6552349894818922765_!_101_!_news_culture_!_你心目中的伟人是谁？_!_
+6552433807960572168_!_101_!_news_culture_!_上联：江中鱼东游西藏，下联怎么对？_!_
+6552436579980280323_!_101_!_news_culture_!_2018艺术厦门博览会5月25日开幕 将展示上万件艺术珍品_!_花鸟画,2018艺术厦门博览会,公共艺术,精神家园,陈立德,艺术家
+6552397951476433166_!_101_!_news_culture_!_上联：草根登上星光道，怎么对下联？_!_
+6552350357035417864_!_101_!_news_culture_!_在历史上或者你身边，有哪些“丁”姓的名字让你念念不忘？_!_
+6552242299076608515_!_101_!_news_culture_!_考场微型小说阅读技法及实例解析_!_蝴蝶,樱子,出人意料,人物性格,写人物
+6552408149846917639_!_102_!_news_entertainment_!_蔡徐坤到底做了什么事，见面会表演要被全程激光笔照，险些失明！_!_蔡徐坤,演唱会,激光笔,见面会,粉丝
+6552458312380580366_!_102_!_news_entertainment_!_黄磊喜得贵子却被批重男轻女 他直言第三胎我更想要女儿_!_黄磊,爸爸去哪,粉丝,一朵金花
+6552141032216068621_!_102_!_news_entertainment_!_52岁的刘嘉玲，62岁的刘晓庆，整容绝对不是最好的抗老秘诀_!_辅酶Q10,武则天,舒适圈,刘嘉玲,刘晓庆,中国电影史,芙蓉镇
+6552336048045687310_!_102_!_news_entertainment_!_最强星声PK戏精之王，新星盛典决赛闪耀苏州！_!_童星,华谊兄弟电影世界,杨佳鹏,说唱歌手,新星盛典苏州分赛区决赛
+6552382266134233604_!_102_!_news_entertainment_!_盘点：拍戏坚决用原声的演员，蒋欣、王凯、靳东、张一山上榜！_!_蒋欣,张一山,片酬,演员,靳东
+6552285218709963271_!_102_!_news_entertainment_!_车工的前世今生——从车工的成长看995的未来_!_
+6552292908823740936_!_102_!_news_entertainment_!_李诞说错话，何炅刘宪华满脸尴尬，张绍刚黑脸：少说话多吃东西！_!_哎哎哎,何炅,说错话,黄磊,张绍刚
+6552459224100635150_!_102_!_news_entertainment_!_陈道明63岁，张国立63岁，而63岁的他看起来像36岁！_!_国家一级演员,中国电影金鸡奖,一九四二,梦驼铃,陈道明,张国立,费玉清,甲方乙方
+6552392586252780035_!_102_!_news_entertainment_!_权志龙，黄景瑜，易烊千玺5月刊-你更喜欢谁的风格？_!_这！就是街舞,吴亦凡,易烊千玺,黄景瑜,男人装,权志龙,陈伟霆
+6552159397357814024_!_102_!_news_entertainment_!_凤姐罗玉凤，靠什么能力在美国生活？_!_
+6552178930323816968_!_102_!_news_entertainment_!_因主角演技“太烂”而放弃的电视剧，若你看完2部，我叫你大神！_!_净天教,上古情歌,韩东君,黄晓明,郑伟文,仙剑奇侠传五,徐惠康
+6552335559396688142_!_102_!_news_entertainment_!_为什么行尸走肉不能看了？_!_
+6552301521965416968_!_102_!_news_entertainment_!_英国脱衣舞娘怀孕9月仍登台 脚踩10厘米高跟鞋_!_脱衣舞,脱衣舞娘,布里斯托尔,拉沃,滑稽歌舞杂剧
+6552342762958619150_!_102_!_news_entertainment_!_他拍戏吓坏孙红雷，拍床戏吓坏对手，为照顾母亲放弃事业四年！_!_石兆琪,战狼,男人养家,盖克,征服,孙红雷
+6552296656631497223_!_102_!_news_entertainment_!_要被真心大冒险这个水晶球男孩笑死了，强行霸道总裁尬到飞起！_!_南京艺术学院,真心大冒险,小姐姐,李东阳,哈哈哈,水晶球
+6552369773508821252_!_102_!_news_entertainment_!_李连杰立遗嘱捐百亿，他拍戏代言没成龙多，为何比成龙还有钱？_!_
+6552442606020723203_!_102_!_news_entertainment_!_抵抗“惯性”很重要，纪录片选题的国外经验先看节展_!_选题,电影节,阿姆斯特丹,圣塞瓦斯蒂安,节展
+6552293333769650696_!_102_!_news_entertainment_!_颖儿肉色礼裙出席活动，网友：透得太惊心动魄，不敢再看了！_!_颖儿肉色礼裙出席,肉色,透得太惊心动魄,颖儿
+6552274888550253069_!_102_!_news_entertainment_!_如果爱：郑晴天是谁扮演的？倪景阳资料背景影视作品是什么？_!_倪景阳,小鱼儿与花无缺,如果爱,女一号,因为遇见你,真情告白,丝丝心动,白日焰火,匹夫,夏至未至,北平往事,大年初一立春,老爸向前冲,出演,金熊奖
+6552376646450544909_!_102_!_news_entertainment_!_如何看待世乒赛女队夺冠后，李晓霞微博霸气送祝福的行为？_!_
+6552438373682774536_!_103_!_news_sports_!_太幸福！詹姆斯绝杀之夜儿子前来观战！赛后父子俩一动作搞笑极了_!_东部决赛,詹姆斯,父子俩,布朗尼,此时此刻
+6552268959750554115_!_103_!_news_sports_!_隆多疑似垫脚库里，勇士或向联盟讨说法，科尔：那是很危险的动作_!_隆多,勇士,科尔,技术犯规,库里
+6552390314227663111_!_103_!_news_sports_!_篮球正负值是什么意思？_!_
+6552470088966996237_!_103_!_news_sports_!_男团决赛，中德男乒大战，马龙兵不血刃3:0赢波尔，为国乒拿到开门红，如何看待？_!_
+6552432757673296397_!_103_!_news_sports_!_谁能娶她？29岁美女奥运冠军何雯娜重回体坛，想找长腿有腹肌男友_!_全运会,蹦床,陈一冰,裁判员,何雯娜
+6552387820353225229_!_103_!_news_sports_!_火箭与爵士的第3战 火箭的轮转发生了变化 这变化将改变一人命运_!_安德森,爵士,火箭,季后赛,乔·约翰逊
+6552405415156515331_!_103_!_news_sports_!_尤纳斯下赛季会留在广东吗？这两人或许比他更合适，朱总会选谁？_!_陈江华,尤纳斯,朱芳雨,杜锋,广东队
+6552450167478419720_!_103_!_news_sports_!_金州勇士、皇家马德里，谁在世界上影响力大？_!_
+6552386912953303300_!_103_!_news_sports_!_如果世乒赛第四局中，刘诗雯不是碰到也同样输一局的平野美宇，结局会怎样？_!_
+6552294478227439879_!_103_!_news_sports_!_虽然中国女乒军团世乒赛3:1战胜日本，但日本球迷说不久日本女乒会赢中国，你怎么看？_!_
+6552135894881534478_!_103_!_news_sports_!_莱万与海因克斯决裂，拜仁还有必要留下他吗？_!_拜仁,海因克斯,莱万
+6552410824906899975_!_103_!_news_sports_!_58同城中甲：替补登场显神功，黄希扬一展皮尔洛神迹_!_两脚球,浙江绿城,黄希扬,中甲联赛,卓尔
+6552464971303748103_!_103_!_news_sports_!_埃梅里：全队在帮内马尔适应巴黎_!_巴黎圣日耳曼,融入球队,内马尔,法甲联赛,埃梅里,SFR体育
+6552362271824151054_!_103_!_news_sports_!_您也辛苦了！泰伦-卢新闻发布会上大口饮水_!_主教练,NBA,泰伦,新闻发布会,nba季后赛
+6552283910703677965_!_103_!_news_sports_!_曼城主帅瓜迪奥拉确认亚亚·图雷今夏离开球队_!_亚亚图雷,科特迪瓦,伊蒂哈德,瓜迪奥拉,布莱顿
+6552323324075049229_!_103_!_news_sports_!_亚洲杯分组国足抽得上上签，国足就一定能出线吗？_!_
+6552290131703759107_!_103_!_news_sports_!_科比他的一生之敌只有一个人，不是詹姆斯，那会是谁？_!_
+6552449920714932487_!_103_!_news_sports_!_2018世界杯有哪些强队缺席？_!_
+6552299340486935053_!_103_!_news_sports_!_荥经县安靖小学开展春季田径运动会_!_田径运动,跳长绳,运动员,荥经县,乒乓球
+6552361909855715853_!_103_!_news_sports_!_厉害了！武汉一高校学子在2018华中区学生跆拳道比赛中夺得14金！_!_跆拳道,武汉大学,中国地质大学,武汉理工大学,武昌理工学院
+6552438196318241288_!_104_!_news_finance_!_得寸进尺！沙特石油还涨！中国霸气拒买，沙特怕了：我是被逼的！_!_中石化,进口量,涨价,原油,伊拉克
+6552466738112365060_!_104_!_news_finance_!_原料上涨叠加需求转旺 纸价开启新一轮上涨_!_瓦楞纸,山东,福建
+6552269277108371971_!_104_!_news_finance_!_产业上下游面对面 服装定制商走访柯桥本土面料企业_!_柯桥,浙江
+6552277047975084558_!_104_!_news_finance_!_“不差钱”的上市公司却偏爱理财 这说明了什么问题？_!_理财产品,格力电器,外延式并购,不差钱,收益率,上市
+6552419674254999822_!_104_!_news_finance_!_在腾讯和阿里刚上市时分别购买100股它们的股票，现在将获利多少倍？_!_
+6552273204461699592_!_104_!_news_finance_!_企业家之殇 上市公司董事长选择自杀真相！_!_周建,绍兴市,金盾股份,民间借贷,企业家
+6552476303692398855_!_104_!_news_finance_!_准备职业炒股该从哪里入手？_!_
+6552298252123767299_!_104_!_news_finance_!_中国芯片第一龙头，横盘2年底部现金叉，5月或超京东方A_!_股市
+6552340327439532552_!_104_!_news_finance_!_重大利好：通策医疗突发重磅信号，下周抄底最佳时机出现！_!_进出口业,医疗服务,通策医疗,医疗,医疗机构
+6552293354770530568_!_104_!_news_finance_!_合伙开店如何转让股份？_!_
+6552329670379962893_!_104_!_news_finance_!_PPP模式聚焦旅游重点领域 社会资本需有更高要求_!_社会资本,母基金,意见,地方政府,PPP模式,旅游项目
+6552296656992207117_!_104_!_news_finance_!_在加拿大生活，消费如何？_!_
+6552276802520220173_!_104_!_news_finance_!_出栏一头猪亏损300元，究竟谁能笑到最后！_!_商品猪,养猪,猪价,仔猪,饲料
+6552483868237103373_!_104_!_news_finance_!_招行、工行、农行、建行和中行五行的信用卡哪个最好用？_!_
+6552480511707578628_!_104_!_news_finance_!_银行卡没有钱，如果一直不用，每年都扣年费等等费用，十年二十年后，银行会主动来要费用吗？_!_
+6552326282024059405_!_104_!_news_finance_!_巴菲特股东大会：关于中国，巴菲特、芒格说了这些_!_投资中国,巴菲特,经济体系,美国运通,奥马哈,芒格,中石化,贸易战
+6552260914303730190_!_104_!_news_finance_!_这真的是一场屠杀，刚刚这个国家崩盘了_!_外汇储备,缩表,阿根廷,外汇管制,阿根廷比索
+6552391117852115204_!_104_!_news_finance_!_为什么是罗森便利店占领了上海的市场，而不是711？_!_
+6552465036814582279_!_106_!_news_house_!_90后女生告诉我：买房比结婚重要_!_女权主义者,女孩子,房产证,安全感,买房子
+6552380751982100740_!_106_!_news_house_!_购房者应该怎样做才能买到称心如意的房产？_!_
+6552412742521716995_!_106_!_news_house_!_300方的地，起6层，打桩起商品房和打桩起厂房，桩有什么不同？_!_
+6552409488731668995_!_106_!_news_house_!_即日起 泸州“1965·茜草工业记忆”博物馆暂停对外开放_!_1965 ·茜草工业记忆,博物馆,泸州文旅集团,泸州,茜草街道
+6552419519367741956_!_106_!_news_house_!_恒大、富力、碧桂园、中梁等地产公司相继进入临汾，你会买哪的房_!_恒大地产,碧桂园,中国恒大,临汾,富力地产,凯隆
+6552298692156588296_!_106_!_news_house_!_物业公司的服务水准非常烂，该怎么解决？_!_
+6552430925467091208_!_106_!_news_house_!_如果2018年房价趋稳，2019年是否会赢来爆发？_!_
+6552361080792809731_!_106_!_news_house_!_为什么会出现“房价跌了反而卖不出去”的现象？_!_
+6552253372785754381_!_106_!_news_house_!_在西安买房哪个位置性价比比较高？_!_
+6552251237599805960_!_106_!_news_house_!_4月楼市成交同比降温 超8城市房地产市场调控加码_!_海南,楼市,杭州,海南省,房价
+6552240030398546179_!_106_!_news_house_!_周口未来房价怎样？_!_
+6552383643287814407_!_106_!_news_house_!_现在买房去售楼部都说卖完了，究竟谁买的呢？_!_
+6552411925995585799_!_106_!_news_house_!_郑州买房除了落户郑州还有其他途径吗？_!_
+6552368087314727181_!_106_!_news_house_!_未来房价会跌到什么程度？_!_
+6552306679898178055_!_106_!_news_house_!_荆门长龙中央公园建小学是真的吗？_!_荆门,中央公园,长龙,楼盘
+6552328146195055111_!_106_!_news_house_!_一周楼市：第六届科博会总体方案确定 城区500余套公租房接受申请_!_科博会,科技城,绵阳,新东方,申请资料
+6521748964591534596_!_106_!_news_house_!_今年以来多地房贷利率上涨_!_家银行,房贷利率,贷款利率,利率上浮,首套房
+6552414030546338056_!_106_!_news_house_!_以后房子会越来越便宜，还是会越来越贵？_!_
+6551396407314481667_!_107_!_news_car_!_聪明男人都会买裸车，再装上这些不贵有档次的车品，开出去倍有面_!_充气床,烟灰缸,密封条,中控台,密封胶
+6552377233216897540_!_107_!_news_car_!_2019最火的豪车，发布一个月订单超37万辆，仅22万比奔驰C级帅！_!_电动汽车,宝马3系,modelS,特斯拉,BBA,奔驰C级,凯迪拉克ATS-L,豪车,凯迪拉克,ATS-L
+6552424891172258062_!_107_!_news_car_!_为什么有些开好车的女生很漂亮？_!_
+6552324743255556615_!_107_!_news_car_!_丰田这次火了！小号“霸道“亮相，2.7L+四驱，25万起后悔买途观_!_SUV,丰田,汉兰达
+6552397680088187143_!_107_!_news_car_!_上班族，25~30w之间选什么车比较好？_!_
+6552356598046523918_!_107_!_news_car_!_买车上牌，还要多花近10万！国内车牌最贵的5个城市_!_车牌,车牌号码,北京,车牌平均价格,上海
+6552307716428136967_!_107_!_news_car_!_路虎揽胜运动——展现雄浑的动感形象_!_路虎揽胜,Ingenium,揽胜运动版,PHEV,路虎揽胜运动,路虎,路虎揽胜运动版,发动机,揽胜
+6552466029946077444_!_107_!_news_car_!_汽车小白的人生第一辆车，名爵6还是马自达昂克赛拉好？_!_
+6552386515555582221_!_107_!_news_car_!_预算15万左右，vv5、马自达cx4和名爵6相比，哪个比较好？_!_
+6552345691975320068_!_107_!_news_car_!_大众怒了，新车完胜雅阁，配2.0T加四驱，仅售12万，还买啥凯美瑞_!_SUV,西雅特,凯美瑞,Ateca,驾驶者
+6552352319063720455_!_107_!_news_car_!_上海国际摩展Honda发布新款电动车_!_CBR1000RR,NS110i,PGM-FI,Honda,摩托车,新大洲本田,电动车,中国消费者,Concept
+6552332115373982221_!_107_!_news_car_!_丰田再次放大招，整车进口，格调不输奥迪A6，售价不足30万要火_!_奔驰C级,A6L,宝马5系,宝马3系,奥迪A6L,THS-II,奔驰E级,英菲尼迪,凯美瑞,TNGA,雷克萨斯ES,凯迪拉克,CT6,A4L,丰田,雷克萨斯,凯迪拉克CT6,奥迪
+6550962018709406212_!_107_!_news_car_!_长5米，比迈腾大一圈，无奈被奥迪A6狂虐，直降7万销量惨淡_!_A6L,奔驰E级,迈腾,凯美瑞,凯迪拉克CT6,奥迪A6L,奥迪,JEEP,蒙迪欧,福特,金牛座,帕萨特,CT6,凯迪拉克,宝马5系
+6552286278858048004_!_107_!_news_car_!_拿到驾驶证后，不敢开车上路的是什么人？_!_考驾照,开车上路,驾驶证
+6552133630779457805_!_107_!_news_car_!_练车时，你被教练骂过吗？怎么骂的？_!_
+6552271577189187847_!_107_!_news_car_!_汽车导航在保修期坏了，自己需要掏钱修吗，还是应该免费？_!_
+6552410529091027470_!_107_!_news_car_!_仅仅7万，标配胎压监测，比哈弗H6还漂亮_!_跨界SUV,哈弗H6,帝豪GS,CX-4,马自达,胎压监测,标配
+6552333653618197000_!_107_!_news_car_!_老婆真的好贴心哟！买这些车载用品安全又享受，驾驶万里都不怕_!_香水,高压水枪,玻璃杯,货到付款,车载蓝牙
+6552459606583410948_!_107_!_news_car_!_针对违章电动车骑车人，让其观看一段时间事故图片及视频，会不会更有效？_!_
+6552447276202066445_!_108_!_news_edu_!_高考加分政策出炉！今年连云港这些高考生可以享受照顾政策_!_技术科目测试合格,文科类,体育类,必修科目,投档
+6552456342659924237_!_108_!_news_edu_!_怎样才能更好的激励自己学习和成长？_!_
+6552449783515054605_!_108_!_news_edu_!_中学年薪50万全国招聘教师，被称为“掠夺师资”_!_教师
+6552468298724803076_!_108_!_news_edu_!_中南财经政法大学分专业招生计划、录取分数_!_分专业招生计划,中南财经政法大学,经济学,对外经济贸易大学,工商管理
+6552478514463899911_!_108_!_news_edu_!_学习工程管理的学生，最适合考建造师和造价师，是不是？_!_
+6552362759701397763_!_108_!_news_edu_!_山东理科生考试平均460分左右，高考会有突破吗？该怎么学习？_!_
+6552252151173759492_!_108_!_news_edu_!_“专升本”录取工作5月下旬进行_!_陕西招生考试信息网,专升本,陕西省,招生办,最低控制分数线
+6552478446738473223_!_108_!_news_edu_!_大学生毕业在即，准备二战考研，但是学校逼着签三方协议，这样合适吗？_!_
+6552099542857155076_!_108_!_news_edu_!_速写，要的就是速度_!_风景速写,速写,秒速写
+6552125196478710275_!_108_!_news_edu_!_冲刺2018年高考数学，典型例题分析63：客观题选讲，提高基础分_!_充分性,y﹣2,x﹣1,解析式,对称轴
+6552271817296314887_!_108_!_news_edu_!_2018年太原市中考体育考试将于5月19日开始_!_四舍五入,太原市,立定跳远,坐位体前屈,选考项目
+6552289567502762499_!_108_!_news_edu_!_考驾照一次性通过，并不一定是技术好！你忽略了最关键的因素！_!_飞行时间
+6551598579599802883_!_108_!_news_edu_!_孩子把别人推下楼后，家长是怎样处理的？_!_熊孩子,家庭教育
+6552417064307392771_!_108_!_news_edu_!_小学生托管老师如何提高检查作业的效率？_!_
+6552417029322703117_!_108_!_news_edu_!_为什么洛伦兹速度变换方程可以推出光速是物体运动的极限速度？_!_
+6552383213338100232_!_108_!_news_edu_!_教育｜4119人！武汉大学自招初审通过人数创纪录_!_武汉大学
+6552286939175715332_!_108_!_news_edu_!_家长为孩子交学费，但却受到教师的轻视，校长说的话让她羞愧不已_!_教师,校长
+6552381294204944648_!_108_!_news_edu_!_招教和特岗教师有什么不同？_!_
+6552379722100113667_!_108_!_news_edu_!_初三的学生马上就要面临中考，有什么好的学习方法推荐吗？_!_
+6552307501226787331_!_109_!_news_tech_!_寒武纪发布新一代AI芯片采用7纳米 华为麒麟980将整入寒武纪IP_!_人工智能,寒武纪,科大讯飞,机器学习,芯片
+6552296410039976205_!_109_!_news_tech_!_电视外接有源音箱有电流噪音怎么解决？_!_
+6552325547379130638_!_109_!_news_tech_!_网上的借贷可以用吗？_!_
+6552252761457557763_!_109_!_news_tech_!_从20%暴跌到0.8%仅用五年，三星手机为何会在中国遇雪崩式溃败？_!_
+6552364767317590532_!_109_!_news_tech_!_创新思考产生创意直至实现产品_!_打印机,好时代,创意很难,网易云音乐,苹果公司
+6552262778244039171_!_109_!_news_tech_!_雷军向李嘉诚展示重磅新机 问了这个问题_!_李嘉诚,智能手机,小米,IDC,策略联盟
+6552459277586399491_!_109_!_news_tech_!_小米上市后市值若超过百度，会带来哪些影响？_!_
+6552258881010008579_!_109_!_news_tech_!_任何指望技术来解决社会公正、公平的设想，都是幻想_!_数据挖掘,大数据,蓄水池,社会公正,格林斯潘
+6552354043484700941_!_109_!_news_tech_!_怎样把自媒体玩得有质量？_!_
+6552425414294241800_!_109_!_news_tech_!_能视频通话，能人脸识别，能语音拍照的8英寸智能音箱_!_空气净化器,音箱,智能家居,京东,加湿器
+6552260300580585742_!_109_!_news_tech_!_自媒体大咖是如何利用自媒体赚钱的？_!_
+6552316594771460365_!_109_!_news_tech_!_国内各类大中型公司正版电脑软件使用率有多高？_!_
+6552155013844566276_!_109_!_news_tech_!_跟万达学管理，怎样保质保量甚至提前完成计划目标？_!_
+6552412344884920584_!_109_!_news_tech_!_1699的魅族m15和1268的魅族pro7标准版该选哪个？_!_
+6552239670879584520_!_109_!_news_tech_!_什么软件应用能下载各种vip视频电影？_!_
+6552310413709541646_!_109_!_news_tech_!_哪个牌子的pos机比较好用？_!_
+6552264830210801934_!_109_!_news_tech_!_朋友圈里的各种信息怎么删除？_!_
+6552402957336314371_!_109_!_news_tech_!_马化腾：从写第一行代码开始，我的理想都是如何做出最好的产品_!_马化腾,微视,腾讯,SNG,刘炽平,任宇昕,腾讯传,腾讯没有梦想,旅行青蛙,恋与制作人,腾讯视频,张小龙,腾讯微博,短视频
+6552272296734621966_!_109_!_news_tech_!_从20%暴跌到0.8%仅用五年，三星手机为何会在中国遇雪崩式溃败？_!_
+6552464349334602254_!_110_!_news_military_!_有一“黑科技”我国已拔头筹，技术具有重要军事价值_!_通讯技术,超级计算机,情报战,数据链,电子对抗
+6552455110851559683_!_110_!_news_military_!_BAR步枪在美军步兵的火力配置中担任什么角色？_!_
+6552397300482703623_!_110_!_news_military_!_跳到水里能躲子弹吗？_!_
+6552338394196738564_!_110_!_news_military_!_为什么战列舰不采用自动装弹，像机枪那样的装置？难办到，不靠谱_!_弹药,机枪,战列舰,保弹板
+6552241996847644931_!_110_!_news_military_!_战略轰炸机的机组人员在执行长航时作战任务中，是怎么吃饭，睡觉和上厕所的？_!_
+6552299206449562125_!_110_!_news_military_!_1.1秒牺牲自己，44毫米拯救4000人，烈士的母亲说：咱娃走得值！_!_生死关头,张德山,跳伞,东方IC,驾驶杆
+6552399632742220302_!_110_!_news_military_!_南海航空兵建设小目标——独立歼灭至少3个尼米兹战斗群！_!_南海,战斗机,海军,防空导弹,海军航空兵
+6552443934679761159_!_110_!_news_military_!_二战德军进攻苏联有没有比巴巴罗萨计划更好的作战方案？_!_
+6552441742342226180_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552264734203183374_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552135788409127171_!_110_!_news_military_!_国外最想买的中国武器装备是哪些？_!_
+6552365851478065411_!_110_!_news_military_!_联合国安理会改革废除一票否决权，会导致什么后果？_!_
+6552445947513667848_!_110_!_news_military_!_F22比J20体积小却重得多，请问F22航程是不是很短？_!_
+6552270570715611400_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552273700320706823_!_110_!_news_military_!_日本强硬向美说不，是真心还是作秀？_!_
+6552465729688437000_!_110_!_news_military_!_历史上的十大军事人才有谁？_!_
+6552339762760385031_!_110_!_news_military_!_全球十大防空导弹盘点：中国只有一款上榜 排第7_!_红旗-9防空导弹,爱国者,S-400,导弹系统,巴拉克,紫菀,S-300V
+6551917967729230339_!_110_!_news_military_!_德国飞行员纷纷选择提前退役，拦都拦不住，只因俄罗斯的疯狂飞行_!_北约,军事实力,德国空军,德国飞行员,战机
+6552455580714271239_!_110_!_news_military_!_叙政府军推进势头太猛，美舰载机猛烈轰炸报复，俄战机起飞对峙_!_大马士革,反政府武装,叙政府军,叙利亚政府军,幼发拉底河
+6552418042377142535_!_110_!_news_military_!_打仗时，钢盔根本不能挡子弹，为何还一定要戴？_!_
+6552252604624142599_!_112_!_news_travel_!_东京游玩8天，如何安排行程？_!_
+6552257845306327560_!_112_!_news_travel_!_人间四月芳菲尽 深山杜鹃始盛开_!_黄山风景区,杜鹃花,大别山,新华网
+6552358557629546760_!_112_!_news_travel_!_川藏线旅游可以选择商务车吗？_!_
+6552424637186179335_!_112_!_news_travel_!_去越南旅行，什么时候去比较合适？_!_
+6552475183930671367_!_112_!_news_travel_!_世界第一伟人是谁？_!_
+6552173874492997891_!_112_!_news_travel_!_一个人坐邮轮旅游有意思吗？_!_
+6552229458634342919_!_112_!_news_travel_!_机票攻略｜说走就走，你还差一张特价机票给你勇气_!_特价机票,廉价航空,航空公司,机票,大航空
+6552423115752734983_!_112_!_news_travel_!_青岛有哪些好玩的地方？_!_
+6552465081064489485_!_112_!_news_travel_!_金寨桂花公园“月亮”之夜美丽多彩！_!_金寨人,金寨,桂花公园,马鬃岭,金寨县
+6552389048529322243_!_112_!_news_travel_!_为什么许多中国人喜好去泰国，而不去三亚？_!_
+6552255393999880455_!_112_!_news_travel_!_宁夏银川影视城好玩吗？_!_
+6552419655196082439_!_112_!_news_travel_!_南宁城市怎么样？适合东北人生活吗？_!_
+6552278403758686723_!_112_!_news_travel_!_第十四届“多浪·龟兹”文化旅游节5月12日开幕_!_温宿天山托木尔文化旅游节,浪·龟兹,旅游节,阿克苏,美食节
+6552448126492344590_!_112_!_news_travel_!_四川游客张某带着六名女子在三亚沙滩烧烤被撒沙，这事你怎么看？_!_
+6552445301087535363_!_112_!_news_travel_!_想去哈尔滨凤凰山旅游，这个地方怎么样？_!_
+6552321503856165134_!_112_!_news_travel_!_四川游客张某带着六名女子在三亚沙滩烧烤被撒沙，这事你怎么看？_!_
+6551668995898999309_!_112_!_news_travel_!_上海这家店生意火爆，大家排着长队只为了吃这道“湖南菜”_!_滴水洞,田子坊,孜然排骨,湖南菜,月桂叶
+6552403355027636743_!_112_!_news_travel_!_夏天太热了，只要有多余的时间很适合去这几个地方漂流，景色清幽_!_孟姜女,连州地下河,黄腾峡,漂流,青龙峡
+6552266837579530499_!_113_!_news_world_!_为什么韩国、越南可以在几十年就把国家文字和语言更改了？_!_
+6552146975343510023_!_113_!_news_world_!_巴以冲突千名巴勒斯坦人受伤 其中包括149名未成年人和8名妇女_!_巴勒斯坦人,示威者,以色列,巴勒斯坦,催泪瓦斯
+6552428561595105800_!_113_!_news_world_!_法国突然宣布重要消息，特朗普大怒，中国却热烈欢迎！_!_法国,特朗普,叙利亚,军工业,马克龙
+6552270338191786500_!_113_!_news_world_!_世界之大无奇不有！Kansas 男子试图与爱车啪啪啪被捕_!_Nissan,Pathfinder,宾夕法尼亚州,电击枪,堪萨斯州
+6552310864135848456_!_113_!_news_world_!_俄罗斯多地爆发集会游行，反对派领袖被捕、一天后获释_!_纳瓦尔尼,普京,俄联邦总统选举法,莫斯科市,莫斯科,俄罗斯
+6552353727242568199_!_113_!_news_world_!_三十年河东三十年河西，美国人寻求的东西，在中国一个都不缺_!_挖泥船,国际空间站,美国,三十年河东三十年河西,主战坦克
+6552385069879656707_!_113_!_news_world_!_如果美国各个州独立了，世界格局会发生什么变化？_!_
+6551006646384984327_!_113_!_news_world_!_世界上哪个国家与美国友好？哪个国家讨厌美国？_!_
+6552352095184355588_!_113_!_news_world_!_美国建国不过两百年，为什么会成为世界强国？_!_
+6552234508945457416_!_113_!_news_world_!_美国建国不过两百年，为什么会成为世界强国？_!_
+6552334651036271108_!_113_!_news_world_!_夏威夷基拉韦厄火山喷发 滚烫熔岩横穿公路_!_夏威夷州,基拉韦厄火山,火山喷发,基拉韦厄,夏威夷
+6552323975546929421_!_113_!_news_world_!_美国建国不过两百年，为什么会成为世界强国？_!_
+6552095951467577860_!_113_!_news_world_!_加拿大男子卡在女厕墙中身亡_!_通风口,加拿大,Diehard,女厕所,卡尔加里
+6552388446797693188_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552295464396390664_!_113_!_news_world_!_日本地铁为什么没有安检？_!_
+6552480711784268040_!_113_!_news_world_!_为什么一个国家不能无限印钞票来让自己有钱，购买更多国外资源以及进行本国建设？_!_
+6552430817379877128_!_113_!_news_world_!_五十退休早吗？_!_
+6552290285890568711_!_113_!_news_world_!_利比亚至今混战，卡扎菲众多女保镖去向成谜，结果超出预知_!_女保镖,美国为首西方国家,利比亚战争,利比亚,卡扎菲
+6552264557052559623_!_113_!_news_world_!_能成为世界第一大国，美国都做对了哪些事情？_!_
+6552295874247000324_!_113_!_news_world_!_如果日本国土滑向海沟沉没，要求中国人民收留，作为中国人你会接受日本人吗？_!_
+6552140663373169166_!_115_!_news_agriculture_!_2018.5.6今日最新猪价_!_110kg
+6552309482725048584_!_115_!_news_agriculture_!_豆腐渣混玉米杆粉发酵变酸能喂猪吗？_!_
+6552241596065120776_!_115_!_news_agriculture_!_保留院落乡土特色避免“千村一貌”_!_中国农业大学,朱启臻,传统文化,农家院落,传统村落
+6552276247395696910_!_115_!_news_agriculture_!_5月中旬，小麦用什么药？_!_
+6552452012305285640_!_115_!_news_agriculture_!_彩礼重，娶妻难！超3000万光棍被判“无妻徒刑”？别急，专家帮您_!_公主嫁到,人口性别比,彩礼
+6552277904980443656_!_115_!_news_agriculture_!_摘樱桃咯！青岛樱桃园大全_!_板栗,滨海大道,樱桃,大樱桃,推出系列果蔬采摘活动
+6552221768432484872_!_115_!_news_agriculture_!_这种农村的美味在哪都能生长，曾经泛滥成灾，现在卖到50元一斤_!_阳荷笋,泛滥成灾,农村
+6552128942784905735_!_115_!_news_agriculture_!_贵州省首个以脱贫攻坚投资基金支持的水利合规性项目获审查通过_!_贵州省,自来水普及率,贞丰县,城乡供水,保证率
+6552277821664788996_!_115_!_news_agriculture_!_如何才把白粉虱、蓟马和红蜘蛛一网打尽？_!_蓟马,保护地,白粉虱,成虫,红蜘蛛
+6552298098712904206_!_115_!_news_agriculture_!_我省突破秸秆生物炭规模化技术，生产成本仅为国外的15%_!_生物炭基肥料,化学肥料,秸秆炭化,规模化,土壤改良剂
+6552268643994960387_!_115_!_news_agriculture_!_补精神之钙 添脱贫动力（话说新农村）_!_贫困地区,顺势而为,羊肚菌,竹荪菌,贫困群众
+6552394244101767687_!_115_!_news_agriculture_!_取消农业户口对农民有哪些好处？_!_养老金,非农业户口,城里人,农业户口,户籍制度
+6552221595765572110_!_115_!_news_agriculture_!_农村老话“家有千万，带毛不算”当你明白道理，为时已晚！_!_养殖业,农村
+6552140837776523783_!_115_!_news_agriculture_!_龙虾养殖：用药不当，可能会导致龙虾大范围死亡_!_养殖,小龙虾,用药,水生动物,生石灰
+6552318033929437700_!_115_!_news_agriculture_!_山东各地抓住小麦孕穗期关键时期防控病虫害_!_鱼台县,小麦,病虫害防治,孕穗期,植保无人机
+6552326687483232781_!_115_!_news_agriculture_!_红薯叶喂猪的诀窍_!_养殖户,红薯,饲料,红薯叶
+6551601141069644039_!_115_!_news_agriculture_!_想回农村开面店，有什么需要注意的吗？_!_
+6552129287900627469_!_115_!_news_agriculture_!_蒜薹两毛钱，说“蒜你狠”的人该闭嘴了_!_山东成武,蒜你狠
+6552386235866808579_!_115_!_news_agriculture_!_专家视点：“优质小麦面积上升，未来小麦价格发展如何”？_!_
+6552315979072798990_!_116_!_news_game_!_绝地求生后座加强导致无法压枪，同时关闭个人交易系统，对此你怎么看？_!_
+6552337677205635588_!_116_!_news_game_!_王者荣耀：张大仙不敢打巅峰赛太菜？抱歉！巅峰赛已经到处是演员_!_玩家,王者荣耀,张大仙,巅峰赛,伪大神
+6552467466402923016_!_116_!_news_game_!_LOL：新一代Soloking史森明，Uzi直播与小明单挑被完虐_!_MSI,Uzi,lol,uzi,SOLO
+6552239203256631555_!_116_!_news_game_!_痴鸡小队的呆鸡有多坑？_!_
+6552191434856857863_!_116_!_news_game_!_《英雄联盟》什么上单英雄能吊打鳄鱼？_!_
+6513152516564189453_!_116_!_news_game_!_炉石传说的卡牌有哪些有趣的外号？_!_二五仔,麻风侏儒,投弹者,吞噬者,水元素
+6552282865046913549_!_116_!_news_game_!_英雄联盟最适合开团四大上单，猴子上榜，第一见人就上_!_猴子,兰博,在上路,对面,机械公敌
+6552461278437179661_!_116_!_news_game_!_AMD出的2200G或2400G玩“吃鸡”的效果如何？需要增加独显吗？_!_
+6552494157242302990_!_116_!_news_game_!_DNF：为什么你达到了打团的要求却没人要？只因为你差了这一件装备_!_超大陆,游戏,dnf
+6552348184474026509_!_116_!_news_game_!_吃鸡再次引发社会事件，背后事件值得深思！_!_社会性,游戏,实锤,CCTV,外挂
+6552283770173522440_!_116_!_news_game_!_作为玩家/赛事观众，你能够接受LOL的版本平衡吗？_!_BAN,版本更新,Wings,Rank,英雄联盟
+6552037868888391949_!_116_!_news_game_!_17shou在Kakao服被封禁！你怎么看待这件事？_!_
+6468789716446609678_!_116_!_news_game_!_游戏剧情里有哪些令人拍案叫绝的「智障桥段」？_!_和平卫士,超级马里奥,诸葛亮,偷马贼,马里奥,文明6,三国战记,巫师3,上古卷轴4,五虎将,合金装备5 ：幻痛
+6552492615122551300_!_116_!_news_game_!_斗鱼黄金大奖赛第四季选拔赛开始，水友点赞热度再创新高_!_FPP,绝地求生,斗鱼直播节,大奖赛第四,黄金大奖赛
+6552318513158029575_!_116_!_news_game_!_在5月5号虎牙索泰杯上，QM战队被实锤代打，如何评价？_!_
+6552269845906325764_!_116_!_news_game_!_手机上有什么好玩的大型游戏？_!_
+6552068748893749767_!_116_!_news_game_!_DNF 天御不闪了接着闪魔战 这两四等一深渊闪的俺们真给跪了_!_dnf
+6551977803959501326_!_116_!_news_game_!_DNF：原来这游戏真的存在狗托，老马居然还对他疯狂暗示！_!_七宗罪,游戏,地下城,dnf
+6552402665832186372_!_116_!_news_game_!_MSI投票榜：Uzi人气爆棚 小花生为非LPL选手第一_!_Mlxg,Peanut,MSI,英雄联盟
+6552292733489250820_!_101_!_news_culture_!_湖湘工匠丨黄彪：“薄施淡染”总相宜_!_茶花,陶瓷,黄彪,陈扬龙,湘字号
+6552313539996942861_!_101_!_news_culture_!_大宋朝才是书法艺术之花盛开的朝代_!_赵匡义,宋太宗,永熙陵,宋朝,书法
+6552372524443762947_!_101_!_news_culture_!_滑县庙会的会本哪里能买到？_!_
+6552293686498034189_!_101_!_news_culture_!_争强不服弱的潘金莲：和小厮闹矛盾，又同母亲吵架，再享鱼水之欢_!_西门庆,潘姥姥,李瓶儿,金瓶梅,可怜天下父母心,潘金莲
+6552487619941368078_!_101_!_news_culture_!_上联：屋后梅花三两亩；求下联怎么对？_!_
+6552449105254154499_!_101_!_news_culture_!_上联：江中鱼东游西藏，下联怎么对？_!_
+6552274693301207566_!_101_!_news_culture_!_道德经第十一章讲解译文繁体小篆版_!_有之以为利,yòng,yǒu,wéi,zhī
+6552456201433514243_!_101_!_news_culture_!_中国古代有休假制度吗？_!_
+6552294788270391566_!_101_!_news_culture_!_下联:夕陽西下已黄昏。上联是什麽？_!_
+6552465322773840141_!_101_!_news_culture_!_上联：往事成空一梦中，下联怎么对？_!_
+6552225490613043460_!_101_!_news_culture_!_上联：寒山寺天寒地冻，下联？_!_
+6552343085370573064_!_101_!_news_culture_!_离愁不是三月雨，下联应该怎么对？_!_
+6552350555392442638_!_101_!_news_culture_!_林徽因拒绝徐志摩而选择梁思成为终身伴侣的原因是什么？_!_
+6552303244041781507_!_101_!_news_culture_!_出联：落英缤纷凌空扬，谁知何处思故乡，如何接联？_!_
+6552374590109122819_!_101_!_news_culture_!_上联：粗茶有真味，下联如何对？_!_
+6552106054195347971_!_101_!_news_culture_!_美术书影：刘继卣绘制的图书插图原稿赏析_!_故事插图,黑旋风李逵,伦勃朗,连环画,刘继卣,武松打虎,鲁智深,七兄弟
+6552454597900763399_!_101_!_news_culture_!_《红楼梦》里，有哪些角色长得像林黛玉？_!_
+6552389013913731598_!_101_!_news_culture_!_开州工匠李文波，一把刻刀写春秋，醉心雕艺铸辉煌_!_木雕,开州区,李文波,创作,美术
+6552321679333261581_!_101_!_news_culture_!_上联：月下品茶悟道，如何对下联？_!_
+6552279699211420163_!_102_!_news_entertainment_!_他出道23年，演技不输孙红雷，因为名字太难让人记住，就没红起来_!_插翅难逃,蒋雯丽,豪哥,赵燕国彰,赵燕国,张子强
+6552396356596531720_!_102_!_news_entertainment_!_马思纯出席活动，网友：天啊，她俩怎么就亲上了_!_马思纯出席活动,马思纯
+6552455203562455556_!_102_!_news_entertainment_!_小龙女失踪半年，曾为见成龙自残，成龙欲秘密见小龙女_!_成龙,吴绮莉,小龙女,监视器,吴卓林
+6552291255332634893_!_102_!_news_entertainment_!_白癜风可以吃虾皮吗？_!_
+6552080137171501576_!_102_!_news_entertainment_!_江疏影红套装现身喜庆惹眼 对镜噘嘴比V心情好_!_
+6552348285049241869_!_102_!_news_entertainment_!_中国人气女明星排行榜2017前十名，谁知道是哪几位？_!_
+6552421539046752772_!_102_!_news_entertainment_!_陈妍希，肉肉的小龙女，你喜欢吗_!_小龙女陈妍希,小龙女,肉肉,陈妍希
+6552372997817106958_!_102_!_news_entertainment_!_好莱坞今日最佳着装_!_ashlee,margot,好莱坞
+6552402570017505795_!_102_!_news_entertainment_!_真人秀综艺新突破，《青春同学会》见证表演界不一样的同学情_!_青春修炼手册,青春同学会,真人秀,中央戏剧学院,上海戏剧学院,同班同学
+6552482153345581581_!_102_!_news_entertainment_!_杨子曾公开表示黄圣依没文化，岳母自爆家门疑为女儿讨公道！_!_扬子,妈妈是超人,北京电影学院,妈妈是超人3,杨子,安迪,黄圣依
+6552472853738947079_!_102_!_news_entertainment_!_张翰公布新恋情？发文：有你的世界，一切才恢复色彩！_!_张钧甯,张翰,古力娜扎,岁岁平安,鹏程万里
+6552491024940270088_!_102_!_news_entertainment_!_如何评价《Fate/Zero》呢？这是我的看法，和你一样吗？_!_圣女贞德,fate/zero,肯尼斯
+6552118234395443719_!_102_!_news_entertainment_!_且行且珍惜：这些明星原谅了出轨的另一半，换个人估计做不到_!_马伊利,白百合,陈思诚,陈羽凡,佟丽娅
+6552286119944258055_!_102_!_news_entertainment_!_幽默笑话，早上起不来床，晚上下不了线_!_还珠格格,幽默笑话
+6552488954103333124_!_102_!_news_entertainment_!_如何评价演员刘芊含的演技？_!_
+6552355550212915726_!_102_!_news_entertainment_!_谢娜复出即获两项“吉尼斯世界纪录”，女神吉尼斯与粉丝难舍关系_!_快乐大本营,Justin,太阳女神,Katy Perry,谢娜
+6552370943564448260_!_102_!_news_entertainment_!_陈冰出席发布会，网友：这二郎腿翘的也太有个性了吧_!_陈冰,陈冰出席发布会,二郎腿翘
+6552375423060148749_!_102_!_news_entertainment_!_空姐与彭于晏亲密合照，喊话他今晚是我的宣示主权_!_主权,空姐,胜利手势,粉丝,彭于晏
+6552419187887702279_!_102_!_news_entertainment_!_《雍正王朝》里赵毅版的李卫演的好还是《李卫当官》里徐铮版的更出色？_!_
+6552390724946493966_!_102_!_news_entertainment_!_瑞士女人花2700万整了一张猫脸，猫看了表示不服_!_女人花,赡养费,上流社会,丛林猫,瑞士
+6552291969895235847_!_103_!_news_sports_!_G3，76人最后时刻疯狂庆祝，为什么反而输球了？_!_
+6552331131725808135_!_103_!_news_sports_!_勇士全神贯注 2017-2018NBA 勇士VS鹈鹕_!_米罗蒂奇,所罗门,库里,克拉克,篮板
+6552313141915549965_!_103_!_news_sports_!_洛瑞缠着詹姆斯，导致詹姆斯倒地差点儿引发冲突，为何要吹詹姆斯犯规？你怎么看？_!_
+6552320323843260680_!_103_!_news_sports_!_如果刘国梁出国执教你是支持还是反对？_!_
+6552410109484466435_!_103_!_news_sports_!_暂停期间隆多亲身向浓眉哥示范如何在内线单打，队友们看着不说话直点头，你怎么看？_!_
+6552238680679907588_!_103_!_news_sports_!_发誓加入中国国籍的黑人拳王阿迪萨亚为什么离开中国了？_!_
+6552303095433396743_!_103_!_news_sports_!_足球茶馆：作为国米球迷你都知道吗？_!_马切吉亚尼,罗纳尔多,联盟杯,拉齐奥,德约卡夫
+6552348791591141636_!_103_!_news_sports_!_世乒赛1/4决赛马龙和许昕均取胜，为樊振东铺平道路，你如何看待？_!_
+6552216606200561923_!_103_!_news_sports_!_大连一方主场3-0击败广州恒大，取得中超首胜，他们取得胜利的原因是什么？_!_
+6552179348621754637_!_103_!_news_sports_!_瑞典世乒赛中国女队夺冠，如何看待新闻发布会上石川佳纯充当中国翻译？_!_
+6552295247508931080_!_103_!_news_sports_!_被打服！骑士詹皇外一人让球迷心宽了，骑士凯尔特人宿敌再见？_!_凯尔特人,猛龙,詹姆斯,骑士,排兵布阵
+6552322177889206536_!_103_!_news_sports_!_詹姆斯8秒绝杀猛龙，洛瑞表情呆滞含有泪花，你怎么看？_!_
+6552401719748198663_!_103_!_news_sports_!_NBA球队是否真的有”相生相克”？_!_
+6552254530992472328_!_103_!_news_sports_!_高山滑雪危险吗？怎么看经常有人摔出赛道？_!_
+6552290870811427080_!_103_!_news_sports_!_如何看待“自由搏击与传统武术谁厉害”这样的问题？_!_
+6552306179647734029_!_103_!_news_sports_!_猛龙为什么又输了？_!_
+6552495300710236430_!_103_!_news_sports_!_为什么说钓鱼穷三代？_!_
+6552482048571867652_!_103_!_news_sports_!_意甲第36轮 切沃2:1力克克罗托内 收获胜利_!_米内罗,比尔萨,切沃,里戈尼,克罗托内
+6552305962672193796_!_103_!_news_sports_!_詹姆斯G3绝杀猛龙，最后时刻追上骑士，猛龙心理有阴影吗？_!_
+6552292809703948558_!_103_!_news_sports_!_世乒赛决赛刘诗雯痛失2赛点，遭伊藤美诚逆转。究竟是哪里出问题了？_!_
+6552430823117685252_!_104_!_news_finance_!_九年首次！伯克希尔一季度亏损，巴菲特却靠苹果派息大赚一笔！_!_巴菲特,伯克希尔,哈撒韦,损益表,苹果公司
+6551937835325194765_!_104_!_news_finance_!_让青春与奋斗同行——海南创业创新青年掠影_!_自然科学,创业创新,海南省,大洲岛,海南
+6552352001429078535_!_104_!_news_finance_!_我在现场：与巴菲特一起寻找下一片投资的热土_!_哈撒韦,股东会,华尔街,巴菲特,投资人
+6552395165074456839_!_104_!_news_finance_!_办信用卡多久可以换工作单位？_!_
+6552498263415587341_!_104_!_news_finance_!_小牛之后，EOS会不会倒下？_!_EOS,ETH,王顺利,柚子,山寨币
+6552487140675027207_!_104_!_news_finance_!_本金120万，融资60万，总资金180万，专职炒股怎么样？_!_
+6552448961519550989_!_104_!_news_finance_!_保险的起源与大航海时代_!_哥伦布,航海,海上保险,大航海时代,西班牙
+6552272529782735111_!_104_!_news_finance_!_如何快速捕捉翻倍牛股，有没有什么技巧？_!_
+6552287088400662792_!_104_!_news_finance_!_合伙开店如何转让股份？_!_
+6552458405380882701_!_104_!_news_finance_!_10个信用卡逾期46万，但每月多多少少都在还，会被报警抓吗？_!_
+6552366193456448004_!_104_!_news_finance_!_苹果或率先破万亿美元？苹果打响反击战 五天暴涨600亿美元_!_iPhone,巴菲特,苹果股票,沃伦巴菲特,苹果市值
+6552364643480764941_!_104_!_news_finance_!_黑龙江最富有的三大家族，首富是陕西人，排名第三的他是黑龙江人_!_誉衡药业,珍宝岛,哈尔滨,方同华,戴永革
+6552303115897405700_!_104_!_news_finance_!_市场比较认可两本理论，《波浪理论》《缠中说缠》大家对这两本理论有什么深入的见解？_!_
+6552343891083788808_!_104_!_news_finance_!_深圳证监局“挥刀”私募新三板 重拳出击违法违规_!_深圳证监局,挂牌公司,新三板,违法违规,深圳
+6552423543970201864_!_104_!_news_finance_!_怎么升级余额宝？_!_
+6552438646861988109_!_104_!_news_finance_!_第五人格哪个监管者好？_!_
+6552350437452808452_!_106_!_news_house_!_个人独资公司购买的房产转到法人名下需要缴纳哪些税费？_!_
+6552352547556819464_!_106_!_news_house_!_夫妻一方偷偷把房子卖了，还能追回房子吗？_!_另一方,违约责任,委托书,购房
+6552358327018324483_!_106_!_news_house_!_买房应该重点看什么_!_房产,容积率,买房,绿化率
+6552451169191461133_!_106_!_news_house_!_土地承包期延长至2057年，农村新增人口如何获得土地？_!_
+6552228424964899080_!_106_!_news_house_!_深圳龙华民治一带到民乐一带，农民房价格大概多少？_!_
+6552325392546398471_!_106_!_news_house_!_从一线城市回太原工作是怎样一种体验？ ​？_!_
+6552479254058107140_!_106_!_news_house_!_2018至2020年房子是涨还是跌？_!_
+6552485069531906318_!_106_!_news_house_!_涨价之前买房的人，现在是什么心情？_!_
+6552446925067518212_!_106_!_news_house_!_一年前花了将近50万买了车库住人，现在不允许住人了该怎么办？_!_
+6552363513170362628_!_106_!_news_house_!_前夫婚前贷款买的房子婚后他加了我名字，离婚之后我再买房子算不算首套住房？_!_
+6552359494683197965_!_106_!_news_house_!_新星宇万城地产双盘联动产品发布会今日举行_!_宇万城,万城公馆,万城院子,万城地产,延吉市
+6552285908115128583_!_106_!_news_house_!_老婆在杭州工作，目前还无房，今年是否是在杭州买房的好时机？_!_
+6552343852500386061_!_106_!_news_house_!_你愿意要清华北大的博士学历，还是北京的一套房子？_!_
+6552345645452099847_!_106_!_news_house_!_无锡这个城市怎么样？_!_
+6552400927142183175_!_106_!_news_house_!_在深圳，作为有钱人是一种什么样的体验？_!_
+6552459965842325773_!_107_!_news_car_!_启辰D60智联精英版分期付款要多少钱？_!_
+6552468755081855236_!_107_!_news_car_!_轮胎侧壁出现裂纹有影响吗？_!_
+6552408164887691524_!_107_!_news_car_!_车跑70空挡N，50再挂D挡跑可以吗？_!_
+6552486408076919048_!_107_!_news_car_!_探界者1.5次低配和速派330舒适怎么选？_!_
+6552151701434925576_!_107_!_news_car_!_想买SUV？这些将要上市的新车绝对有适合你的_!_SUV车型,一汽,紧凑型SUV,哈弗,D80,一汽奔腾
+6552482725427675400_!_107_!_news_car_!_遇到抢劫，但在他们没注意的情况下跑了，他们在追的过程中被路过的汽车撞死了，如何进行责任划分？_!_
+6552468036660494599_!_107_!_news_car_!_雅阁，凯美瑞和天籁的后期费用，大约都是多少钱？_!_
+6552299160190583303_!_107_!_news_car_!_小轿车底盘生锈怎么办？教你一个好方法，5分钟除锈还防刮花_!_底盘,小轿车,底盘护甲
+6552303709940875784_!_107_!_news_car_!_「陕西传媒网」西安机动车达300万辆 专家：道路规划影响拥堵_!_机动车,停车费,公共交通,拥堵费,保有量,机动车保有量,西安市
+6552492310460891656_!_107_!_news_car_!_买宝骏730的车主后死悔！又一黑马SUV来袭，车长4米8，仅6万起！_!_宝骏730,汉腾,SUV,中控台,五菱宏光,宝骏
+6552271653932368142_!_107_!_news_car_!_怎么看小车轮胎的日期？_!_
+6552355709550330382_!_107_!_news_car_!_安全倒车9不要！_!_天津
+6552323932513370627_!_107_!_news_car_!_今年起亚将会大反扑，三款车型可与国产车抗衡，最后一款势不可挡_!_韩系车,CROSS,智跑,K3,东风悦达起亚,起亚
+6552430602744758787_!_107_!_news_car_!_号称最安全的中型SUV，性能不输奔驰GLC和宝马X3，才36万元开回家_!_奔驰GLC,X3,沃尔沃XC60,XC60,宝马X3,沃尔沃,显示屏,SUV
+6552310950513345027_!_107_!_news_car_!_要是朗逸plus只要12万起步，你还会买朗逸？_!_朗逸Plus,途观,紧凑型轿车,帕萨特,朗逸,大众朗逸,途观L,上汽大众
+6552490590276157955_!_107_!_news_car_!_汽车别直接去镀晶了，越镀越伤车漆！教你一新招，天天开干净新车_!_洗车,降温剂,车漆,挡泥板,汽车底盘
+6552418719048401155_!_107_!_news_car_!_新能源汽车有发展前景吗？_!_
+6552270978133525006_!_107_!_news_car_!_2017款福特猛禽F150 白金版顶配夏季爆款_!_福特,F150,福特猛禽,皮卡,MyFord,液晶屏,福特F150
+6552273356782043656_!_107_!_news_car_!_轿车突然冲向马路中央撞翻 路人打开车门救人却 傻眼了_!_本田雅阁
+6552386973573579268_!_108_!_news_edu_!_浙江外国语学院_!_校际合作框架协议,北京外国语大学,教育学,商务英语,西班牙语,语言文学
+6552375982148289038_!_108_!_news_edu_!_第三届全国基础教育信息化应用展示交流活动在京举行_!_教育司,创意彩泥粤剧体验平台,参观者,基础教育信息化,新疆生产建设兵团
+6552430362734100744_!_108_!_news_edu_!_天津工业大学和南昌大学，哪个好？_!_
+6552240684307317261_!_108_!_news_edu_!_“北大校长错别字”背后折射的现象令人深思_!_林建华,错别字,陈涉世家,校长,北京大学
+6552383922322276612_!_108_!_news_edu_!_如何在亏损中学习、进步、成功？_!_
+6552487349098381838_!_108_!_news_edu_!_照亮我人生的东辉老师，您在哪里？︱等着我寻人_!_洮南,东辉,好想你,教师
+6552321373987930376_!_108_!_news_edu_!_流浪狗扰乱正常教学秩序，破坏学生们生活，该怎么办？_!_
+6552294536876392973_!_108_!_news_edu_!_每个大学后面都有一个大佬，你们猜是谁？_!_
+6552283900880617987_!_108_!_news_edu_!_小升初17年南京真题，据说题目类型没有多大变化，尽早看看！_!_小升初,南京
+6552355524967399943_!_108_!_news_edu_!_1~6年级数学公式表，悄悄打印！给孩子挂墙上背，清华北大随便考_!_基础知识,数学
+6552490730424631812_!_108_!_news_edu_!_一年级下册数学期中试卷，有点难，尖子生也考不了满分_!_期中考试,年级,尖子生
+6552457349141889540_!_108_!_news_edu_!_从出题套路中掌握考高分技巧，揭秘中考数学出题套路（三）_!_填空压轴题型,填空题压轴题型,全等三角形,出题套路,三角形,相似三角形,勾股定理,天龙八部
+6552373518347010567_!_108_!_news_edu_!_高考“白卷”考生10年后考取大专 后悔浪费高考机会_!_高考,幡然悔悟,教育宣言,专升本,徐孟南
+6552276146564629006_!_108_!_news_edu_!_不算吃亏（民间故事）_!_登山,这群人,王更豪,澳洲,闺女,王土豪,何娟,托福考试,张燕燕,钢琴
+6552370970554794509_!_108_!_news_edu_!_这两种极端背单词方法你会选哪种？后一种效果没你想的那么好_!_单词,笔记本,背单词
+6552323167711396104_!_108_!_news_edu_!_考上中国矿业大学北京研究生是什么感觉？_!_
+6552472552436924676_!_108_!_news_edu_!_高考只剩一个月，应该怎样复习？_!_
+6552437730117157380_!_108_!_news_edu_!_专家点评2018年高中学业水平考物理等级考试卷（附参考答案回_!_原子物理,思维能力,试题,欧姆定律,物理
+6552400428040978692_!_108_!_news_edu_!_NSAC的证书是认可的吗？为什么在国家证书查询中查询不到？_!_
+6552450337859437060_!_108_!_news_edu_!_4000多观众齐聚镇雄实验中学，五月文化节文艺汇演嗨翻全场_!_爱在天地间,打跳欢歌,葫芦丝,赤水河边,文艺汇演,田径运动,镇雄,逆战,实验中学
+6552462013207937550_!_109_!_news_tech_!_虚拟货币-从EOS主网上线，巴菲特不看好虚拟货币解读虚拟货币_!_比特币,区块链,EOS,虚拟货币,ICO
+6552127247279456772_!_109_!_news_tech_!_2018年款iPad怎么样？用过的人还会选择安卓吗？_!_iPad,平板电脑,安卓平板,iPad Pro,iPadmini
+6552310583578853640_!_109_!_news_tech_!_持股比例仅次于雷军，如何评价小米二号人物林斌？_!_
+6552372359121076749_!_109_!_news_tech_!_店铺分销模式，分销宝智能化采集让你出单变得简单_!_阿里巴巴,分销,聚划算,互联网,店铺
+6552306325353660675_!_109_!_news_tech_!_如果所有银行不让支付宝绑定银行卡了，支付宝会怎样？_!_
+6552467747576480004_!_109_!_news_tech_!_一个人说他喜欢你，心里有你，可是从来没打过电话，只是微信聊天，他是真心的吗？_!_
+6552313123737436430_!_109_!_news_tech_!_蚂蚁花呗是否属于小额贷款，会影响个人征信吗？_!_
+6552465024961478915_!_109_!_news_tech_!_有什么办法可以下载酷狗里面的收费歌曲呢？_!_
+6552292003093152008_!_109_!_news_tech_!_把支付宝里的收款码发给别人安全吗？_!_
+6552466374684312077_!_109_!_news_tech_!_关于《腾讯没有梦想》_!_腾讯没有梦想,马化腾,腾讯
+6552258232436392455_!_109_!_news_tech_!_这一年，阿里巴巴要怎么打破“诅咒”，实现马云的新零售大梦_!_马云,新零售,天猫,阿里,Lazada
+6552353950538924301_!_109_!_news_tech_!_如果所有银行不让支付宝绑定银行卡了，支付宝会怎样？_!_
+6552268545688862980_!_109_!_news_tech_!_2200荣耀V10与2599荣耀10应该怎么选？_!_
+6552470575283962120_!_109_!_news_tech_!_下班后，有什么可以在网上做的兼职？_!_
+6552336574674108931_!_109_!_news_tech_!_手机挖矿app，数字币AIC、SKU_!_填写邀请码,AIC,区块链,SKU
+6551688826631225614_!_109_!_news_tech_!_Instagram上线支付功能可在APP内直接购物，为什么社交软件巨头纷纷布局电商？_!_
+6552350316371640583_!_109_!_news_tech_!_从20%暴跌到0.8%仅用五年，三星手机为何会在中国遇雪崩式溃败？_!_
+6552380036815520003_!_109_!_news_tech_!_阿里巴巴最大股东是日本银软吗？_!_
+6552284077142049032_!_109_!_news_tech_!_华为若被美国禁了芯片和系统，影响大吗？_!_
+6552392111348515080_!_110_!_news_military_!_美国为什么害怕自己出售的武器被用来对付自己？_!_
+6552373438223221251_!_110_!_news_military_!_狙击手为什么要在射击之前将第一枚子弹给退掉？90%的人都想不到_!_枚子弹,狙击手
+6552274394087948551_!_110_!_news_military_!_冷战期间，美国和苏联的坦克差异为什么那么大？_!_
+6552305453722763533_!_110_!_news_military_!_联合国安理会改革废除一票否决权，会导致什么后果？_!_
+6552245217804681486_!_110_!_news_military_!_辽宁舰航母的斜角甲板是14度，国产航母的斜角甲板是12度，两度之差有何不同？_!_
+6552330446791770372_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552254238318133518_!_110_!_news_military_!_为何美国从未发生过军人独裁统治？_!_
+6552363928460984836_!_110_!_news_military_!_长五总装团队：正值芳华肩负新使命 我为祖国铸利“箭”_!_科技集团,长征五号,整流罩,天津火箭公司总装车间,运载火箭
+6552306111284773128_!_110_!_news_military_!_当年美国向日本扔原子弹的地方，现在怎么样了？_!_
+6552111316411089422_!_110_!_news_military_!_对越反击战，他1人毙敌18人，同事与他相处10年，才知他真实身份_!_反击战,杨启良,边防部队,第1军,越军
+6552124673847460100_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552458616555700487_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552322435494969603_!_110_!_news_military_!_联合国安理会改革废除一票否决权，会导致什么后果？_!_
+6552474271849906701_!_110_!_news_military_!_太彪悍！解放军某炮兵连惊现上等兵“副连长”_!_上等兵,小齐,炮兵连,过得硬,上等兵副连长
+6552273732893671683_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552488340120142349_!_110_!_news_military_!_运气太差？俄上苏30战机地中海坠毁原因成谜！一细节待解_!_战机,苏霍伊,苏30,坠毁,苏27,苏35,A-18
+6552339800639144451_!_110_!_news_military_!_全球十大防空导弹盘点：中国只有一款上榜 排第7_!_S-300V,红旗-9防空导弹,爱国者,S-400,导弹系统,巴拉克,紫菀
+6552324968623898883_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552470840758239752_!_110_!_news_military_!_美航母空袭之后 大批美特种兵参战：叙利亚爆发新战争_!_杜鲁门,美军,叙利亚,特种兵,库尔德
+6552241692106293763_!_110_!_news_military_!_“科摩多-2018”多国联合演习开幕_!_舰载直升机,龙目岛,国海军,中国海军,印度尼西亚海军
+6552358423617339912_!_112_!_news_travel_!_旅游｜宜兴人为什么都偏爱古镇_!_锡帮菜,古镇,水墨画,唐伯虎,荡口古镇
+6552474960273604868_!_112_!_news_travel_!_世界第一伟人是谁？_!_
+6552366462919508484_!_112_!_news_travel_!_越来越多女人出门带上这些东西，外出旅游更放心_!_氢离子,反向伞,望远镜,便携式,睡袋
+6552429266393366788_!_112_!_news_travel_!_青山绿水依然在，下联如何对？_!_
+6552497609326461191_!_112_!_news_travel_!_宝安石岩卖冰鲜海产去哪里进货比较合适？_!_
+6552451326133928461_!_112_!_news_travel_!_全域体育助力全域旅游 全民健身助推大健康产业_!_毕节市,山地自行车,健身,百里杜鹃宣传部供图,奢香
+6552326018416247303_!_112_!_news_travel_!_浦口火车站的沧桑记忆_!_梧桐树,津浦铁路,相思树,锁麟囊,浦口火车站,火车站
+6552142438780109316_!_112_!_news_travel_!_埃米尔冈公园：各色郁金香争相斗艳，徜徉在花海又是怎样的感觉？_!_郁金香属,郁金香节,郁金香,东方IC,土耳其
+6551268868336124423_!_112_!_news_travel_!_新疆最美的7个地方，你都知道是哪里吗_!_果子沟,喀纳斯湖,赛里木湖,纳木错,巴音布鲁克
+6552458467733406221_!_112_!_news_travel_!_国内这座古城投资20亿，不收门票却游客稀少，当地人都不愿意去_!_龙潭水乡,成都,自然资源,古城,江南水乡
+6552280160442253575_!_112_!_news_travel_!_去长白山玩，有什么特产值得买？_!_
+6552436093352935944_!_112_!_news_travel_!_假期里，别样的旅行！_!_四川,新农村,导游
+6552364189896147470_!_112_!_news_travel_!_有些爱，我想邮给您！“十年·邮爱”主题活动正式启动_!_汶川大地震,四川新闻网,绵竹市,十年·邮爱,绵竹
+6552399296304513539_!_112_!_news_travel_!_中国游客到北海道旅游，“冒死”吃帝王蟹 吃到扶墙而去！_!_函馆朝市,帝王蟹,北海道,皇帝蟹,海鲜
+6552403590739132932_!_112_!_news_travel_!_走进红色文化旅游胜地 领略路集千年古会风情_!_纪王场乡,利辛县,李世民,三月二十一,路集
+6552456504383898120_!_112_!_news_travel_!_中国唯一的俄罗斯民族乡，归属于内蒙古，却住着30%俄罗斯族人_!_小镇,俄罗斯族,额尔古纳河,俄罗斯,大兴安岭
+6552498162001510669_!_112_!_news_travel_!_为什么广西北海地少人少，经济总量却比南宁高？_!_
+6552342259277234696_!_112_!_news_travel_!_四川九寨沟，美过童话的山水“九寨归来不看水”世界自然遗产名录_!_长海,世界自然遗产名录,五花海,九寨沟县,九寨沟,孔雀河
+6552233262519943431_!_113_!_news_world_!_英拉去哪里了？_!_
+6552164070156402951_!_113_!_news_world_!_季莫申科目前民调支持率最高，她有机会成为下一任乌克兰总统吗，你怎么看？_!_
+6552368829433905416_!_113_!_news_world_!_作为俄罗斯总统，普京上台以来有哪些失误？_!_
+6552252195939549448_!_113_!_news_world_!_如果证实反政府武装使用化学武器，美英法会被追责吗？_!_
+6552432921263735048_!_113_!_news_world_!_印度一男子不顾劝阻要和熊自拍，被咬死，大家怎么看？_!_
+6552359904793854215_!_113_!_news_world_!_美国超级英雄是怎么来的？_!_
+6552443265235288324_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6551282307917938948_!_113_!_news_world_!_如何看待日本的发展？_!_
+6552480731963064579_!_113_!_news_world_!_中国禁止进口洋垃圾，为什么英国很慌张？_!_
+6552392907540660488_!_113_!_news_world_!_如果完全按以色列的思路，巴以问题将如何解决？_!_
+6552202211198238989_!_113_!_news_world_!_美国建国不过两百年，为什么会成为世界强国？_!_
+6552359214784708868_!_113_!_news_world_!_美国建国不过两百年，为什么会成为世界强国？_!_
+6552235869615423752_!_113_!_news_world_!_美国法院判伊朗赔付9·11遇难者家属逾60亿美元，你怎么看？_!_
+6552365741344030980_!_113_!_news_world_!_美国现在有几位前总统健在？_!_
+6552458358022996228_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552138777593118979_!_113_!_news_world_!_普京在刚上台时说要带领俄罗斯成为强国，就目前的形势来看，俄罗斯是否成为了强国？_!_
+6552380612601184520_!_113_!_news_world_!_日本和以色列为什么会在贫瘠的土地上建立起高度发达的国家？_!_
+6552476374676799757_!_113_!_news_world_!_日本的自杀率为什么这么高？_!_
+6552417809521967363_!_113_!_news_world_!_一战末期，列宁不惜割让大片国土求和，是否合理？_!_
+6552371191175184904_!_115_!_news_agriculture_!_如果你有50万，是愿意在农村盖别墅，还是去小县城买一套小房子？_!_县城买,农民兄弟,农村,租房子,县城
+6552376900470178317_!_115_!_news_agriculture_!_唐培科：带领村民们打造“新农村”_!_劳动模范,唐培科,腰站子村,新农村,合作社
+6552299116771148291_!_115_!_news_agriculture_!_小小说：李书记卖鸭_!_鸭子,香港,李书记,胡子拉碴,县委书记
+6552399786027254280_!_115_!_news_agriculture_!_老农去摘玉米，却在玉米秆上捡到很多现金，让人哭笑不得_!_玉米,农家乐,玉米秆
+6552419760498278926_!_115_!_news_agriculture_!_4000家农资店齐努力，农资拒赊风暴无人能阻！_!_农资经销店,肥料圈,农资,经销店,买化肥
+6552284683432886797_!_115_!_news_agriculture_!_养殖户希望猪价再跌一跌，农村散养户和规模化猪场谁能笑到最后！_!_养猪,养殖户,规模化,猪场,母猪
+6552163838177837316_!_115_!_news_agriculture_!_农村俗语：“人穷断六亲”，这句话真实吗？为什么？_!_
+6552269016428184072_!_115_!_news_agriculture_!_农民在地里三件事情不能随意的做，不然可能会失去土地！_!_修建房屋,农业,农村
+6552331556478779908_!_115_!_news_agriculture_!_秦绪文：玩抖音怎么才能赚到钱？_!_PIPA,今日头条,抖音,养殖业,罗振宇
+6552275816720368132_!_115_!_news_agriculture_!_黑龙江抚远开展最大规模大马哈鱼放流行动_!_北太平洋,抚远市,大马哈鱼,黑龙江,乌苏里江
+6552331970330755587_!_115_!_news_agriculture_!_为了寻找小时候的记忆，一大早往老家赶，找的就是这种酱果树莓_!_小白花,杨梅树
+6552304868076290564_!_115_!_news_agriculture_!_农村酒席上必不可少的东西，都在这里了！_!_四季豆,农村,湖南
+6552320582329827598_!_115_!_news_agriculture_!_现在农村人为什么都不种地了？_!_
+6552460972294930947_!_115_!_news_agriculture_!_在延津有一种聊天叫喷空儿_!_喷空儿,生产队,席地而坐,马庄乡,田边村
+6552295387846148611_!_115_!_news_agriculture_!_华池县马铃薯产业有序发展_!_示范点,马铃薯,华池县,贫困户,新品种
+6552378249442230797_!_115_!_news_agriculture_!_武汉再提城乡低保标准 城市每人每月提高到715元_!_农村,特困人员,武汉,孤儿供养标准,武汉市
+6552419254447112717_!_115_!_news_agriculture_!_连饭都吃不上_!_
+6552497859994845444_!_116_!_news_game_!_dnf旭旭宝宝红眼新高度百万战力，五件娜迦王海伯伦加持，和国服其他红眼差距有多大？_!_
+6552481216531005966_!_116_!_news_game_!_官总一个职业选手被路人大神说是总导演？这得花多少钱？_!_英雄联盟,实名制,打上国服第一,LOL,LDL
+6552303885728350723_!_116_!_news_game_!_两个在各自的游戏上称得上一哥的主播，为什么到最后下场却不一样_!_卢本伟,DNF,赵梦玥,游戏
+6552408147577798920_!_116_!_news_game_!_你认为红白机历史上最难的三款游戏分别是什么？_!_
+6552302276654924045_!_116_!_news_game_!_为什么有些斗鱼的主播们这么喜欢拿观众当傻子一样耍？_!_
+6552480379805106695_!_116_!_news_game_!_震惊！LPL冠军下路首秀惨遭黄金奶妈双杀，Uzi：现在黄金都这么强_!_uzi,LPL冠军,Uzi,MSI
+6552358825788178695_!_116_!_news_game_!_手游哪些最好玩？_!_
+6552459381093433864_!_116_!_news_game_!_一个梦幻平民老玩家的商人之路！——5_!_游戏
+6552183924028604935_!_116_!_news_game_!_如果禁止未成年人充值腾讯游戏，腾讯营收会受到多大影响？_!_腾讯,游戏,未成年人,小学生,腾讯游戏
+6552295272913830414_!_116_!_news_game_!_王者荣耀：辅助开局这么做，准备6分钟投降吧！他绝对是个坑_!_打野,游戏
+6552156878594376195_!_116_!_news_game_!_世界上最大的格斗游戏比赛EVO，街霸5报名人数仅排第二_!_格斗游戏,贝吉特,EVO,街霸,任天堂大乱斗
+6552375157745254916_!_116_!_news_game_!_《荒野大镖客2》细节整理 支线、狩猎、荣誉系统_!_玩家,GTA5,散弹枪,西部片,马匹,荒野大镖客2
+6552372203856331272_!_116_!_news_game_!_「王者荣耀」S11最强上单是他，梦琪和他有一拼_!_王者荣耀,s11,程咬金,游戏
+6552491312749216260_!_116_!_news_game_!_爆款RPG《东方梦符祭》爽快抽卡停不下来_!_博丽灵梦,荒神罪,乱战先锋,东方梦符祭,莉莉白,卡牌,SSR,DOTA2,灵梦
+6552461153572749576_!_116_!_news_game_!_《炉石传说》现在不花钱还能有较好的游戏体验吗？_!_
+6552290581672886797_!_116_!_news_game_!_绝地求生里，能用赏心悦目来形容他们操作的主播，他当榜首_!_LOL,出神入化,绝地求生,游戏,主播
+6552364656223060231_!_116_!_news_game_!_怎么样能直播迷你世界？_!_
+6552483893633614088_!_116_!_news_game_!_你第一次被魔兽世界震撼到是什么时候？_!_
+6552289200924787214_!_116_!_news_game_!_王者荣耀KPL：泥潭深入的QG和已经触底的AG，谁更值得同情？_!_QGhappy,eStar,季后赛,AG超玩会,KPL
+6552223105224278275_!_101_!_news_culture_!_应该如何评价硬笔书法名家范林庆的书法水准？_!_
+6552340581471748360_!_101_!_news_culture_!_深夜无眠谁共语，如何对下联？_!_
+6552331407862006024_!_101_!_news_culture_!_上联：梦里水边情，回味绵长，下联怎么对？_!_
+6552446523618099720_!_101_!_news_culture_!_中国书法博大精深，见字如见人，是时候练练自己的字了_!_书法家,字帖,书法,文房四宝,字如其人
+6552496532128530701_!_101_!_news_culture_!_上联：女子好，如何对下联？_!_
+6552407773315858696_!_101_!_news_culture_!_下联：饱暖思淫欲。请赐上联？_!_
+6552288898263810311_!_101_!_news_culture_!_明末出现那么多名将，大明为什么还是亡了？_!_
+6552401468450668808_!_101_!_news_culture_!_云横东岭松挺立；有什么好的下联呢？_!_
+6552401483411751171_!_101_!_news_culture_!_为什么大家都不喜欢于丹了，从热捧到抛弃经历了什么？_!_
+6552387345734172935_!_101_!_news_culture_!_上联：江中鱼东游西藏，下联怎么对？_!_
+6552425266918981902_!_101_!_news_culture_!_日本天皇有“三大神器”是草薙剑、八咫镜和八坂琼曲玉，传说威力十足，为何一直不愿公诸于世？_!_
+6552436282931282446_!_101_!_news_culture_!_孙悟空真身竟然是个皇帝？难怪《西游记》被禁！_!_吴承恩,孙悟空,傲来国,大洪山,嘉靖皇帝,西游记,勘破西游
+6552368343683170573_!_101_!_news_culture_!_上句：浊水三江一夜雨，如何接句？_!_
+6552378503537361160_!_101_!_news_culture_!_故宫断虹桥那只“护档狮”有什么来历？_!_
+6552304220500918531_!_101_!_news_culture_!_“全聚德”的“德”为什么少了一横？_!_
+6552434771710968072_!_101_!_news_culture_!_上联：庭前绿竹迎风舞，下联应该怎么对？_!_
+6552290230076965127_!_101_!_news_culture_!_你怎么看待用书法创业？_!_
+6552452053090697476_!_101_!_news_culture_!_上联：半床明月增寂寞，下联怎么对？_!_
+6552413720641798413_!_101_!_news_culture_!_怎么用六个词语概括自己的故事？_!_
+6552482554237157646_!_101_!_news_culture_!_《儒林外史》里哪个人物最无耻？_!_
+6552321866231448068_!_102_!_news_entertainment_!_《爸爸去哪儿6》嘉宾确定，老爸团颜值超高，他却拖了团队的后腿_!_妈妈是超人3,妈妈是超人4,爸爸去哪儿6,张丹峰,包贝尔,王力宏,亲爱客栈,李承铉,张晞彤
+6552260195743957512_!_102_!_news_entertainment_!_中国故事大会第二季冯提莫是第几期，冯提莫1500万粉丝背后的故事_!_黑豹乐队,中国故事大会,音乐精灵,冯提莫,曹景行,网络主播
+6552362698472948237_!_102_!_news_entertainment_!_被遗忘的10部国产良心剧，演员阵容强大，多个国家一级演员出演_!_丁嘉丽,山城棒棒军,福贵,王志飞,马少骅
+6550783534808695309_!_102_!_news_entertainment_!_论饰演妲己，林心如远不如范冰冰，但是最经典的还要数这个人_!_封神榜,莲花童子哪吒,林心如,张馨予,李依晓,傅艺伟,妲己
+6552429702508708360_!_102_!_news_entertainment_!_《金粉世家》张琼姿女儿长期遭霸凌，如今身材大好，美艳成仙！_!_台湾,夜市人生,模特,金粉世家,职业棒球,张琼姿
+6552377911326802439_!_102_!_news_entertainment_!_贝克汉姆吴秀波金城武小李子，这几位帅哥大叔倾国之貌_!_莱昂纳多·迪卡普里奥,贝克汉姆,枪王之王,吴秀波,东方IC,金城武
+6552452106698097166_!_102_!_news_entertainment_!_「吃鸡竞技」系列文，大吉大利，今晚吃鸡！_!_处处吻
+6552496204314313219_!_102_!_news_entertainment_!_倪妮：优雅大方_!_
+6552274147823583747_!_102_!_news_entertainment_!_吴亦凡出席活动和某女星互动，网友：看他们的坐姿好像很不自在_!_
+6552447270715916803_!_102_!_news_entertainment_!_杨洋就是我们的皮系网管“皮皮修”没有错了！_!_全职高手,武动乾坤,皮皮修
+6552341601450983944_!_102_!_news_entertainment_!_美国女星凯莉·詹娜黑色纱裙参加好莱坞派对，造型时尚_!_凯莉·詹娜,女星,美国,凯莉,好莱坞
+6552432197662409220_!_102_!_news_entertainment_!_甄嬛传：凡是和甄嬛对着干的人，结局都不好，为啥他却安然无恙！_!_沈眉庄,甄嬛,太医院,皇后,华妃
+6552321918744133892_!_102_!_news_entertainment_!_台湾的“槟榔西施”很常见吗？_!_
+6552254437270749710_!_102_!_news_entertainment_!_林心如现身活动大秀香肩，分享女儿小海豚近况：会叫爸爸妈妈了_!_恋爱结婚,霍建华,林心如,连体裤,海豚
+6552481295283257864_!_102_!_news_entertainment_!_谢娜《快本》首秀令人感动，生日之际《快本》是最大的礼物！_!_百褶裙,吴昕,大本营,快本,谢娜,双喜临门
+6552476215855284740_!_102_!_news_entertainment_!_泰剧圈已被这4位高颜值实力小花旦占领了，Bella有赶超Yaya的势头_!_Weir,逐爱天涯,鬼夫,娱乐圈,恋爱诊疗中,花旦,恶玫瑰公主,龙血翡翠,爱与罚,马里奥,魔法床2017,舞之殇,泰国,天生一对,名门绅士之珍爱妙方
+6552415116384535053_!_102_!_news_entertainment_!_《后西游记》之10巧计救太后（浙江人美总17册）_!_
+6552471937619067395_!_102_!_news_entertainment_!_追星追出新高度！浙大医学生SCI论文致谢林俊杰_!_胡江华,The,浙江大学,林俊杰,SCI论文
+6552422460719890952_!_102_!_news_entertainment_!_撩谢霆锋？张柏芝“花痴”上身，自曝喜欢黄磊：既会做饭又有爱心_!_黄磊,花痴,谢霆锋,妈妈咪呀,如果，爱,张柏芝
+6552364163794993678_!_102_!_news_entertainment_!_张天爱出席发布会，网友：葫芦臀实在太抢镜了，羡慕这样的身材_!_身材张天爱出席发布会,张天爱,葫芦臀实在太抢镜,葫芦臀
+6552361067085824520_!_103_!_news_sports_!_师徒之争 舒斯特尔给卡纳瓦罗上了一课_!_卡纳瓦罗,U23政策,恒大,卡帅,舒斯特尔
+6552157903388672264_!_103_!_news_sports_!_今年东决如果骑士打凯尔特人，怎么样？_!_
+6552263488893354247_!_103_!_news_sports_!_昔日女乒主帅施之皓，最近情况怎么样啊？_!_
+6552334410572628483_!_103_!_news_sports_!_25球！国家德比的头号射手——梅西_!_西甲皇马-巴萨,梅西,皇马,超级杯巴萨-皇马,巴萨
+6552436469938520328_!_103_!_news_sports_!_为什么女子网球服下装是裙子？_!_
+6552440136808792327_!_103_!_news_sports_!_詹姆斯再次绝杀，3-0领先猛龙，你怎么看？_!_
+6552289899507089933_!_103_!_news_sports_!_「赛事预测」莱加内斯 VS 莱万特_!_Levante,马拉加,Malaga,莱万特,莱加内斯
+6551834697796682243_!_103_!_news_sports_!_穆里尼奥希望桑德罗来代替卢克肖在曼联的位置_!_桑德罗,曼联,穆里尼奥,左后卫,阿莱克斯
+6552430684755984900_!_103_!_news_sports_!_不要命了？生死边缘还敢送温暖！热刺托大后果严重_!_西布朗,热刺,切尔西,利物浦,凯恩
+6552452395102634254_!_103_!_news_sports_!_瑞典世乒赛中国女队夺冠，如何看待新闻发布会上石川佳纯充当中国翻译？_!_
+6552267169558692100_!_103_!_news_sports_!_周琦为塔克唱生日歌，全队被逗的哈哈大笑，你怎么评价？_!_
+6552264965707792910_!_103_!_news_sports_!_请许老板打开支票本为恒大引入球员，有些人留着真没什么用！_!_广州恒大,王上源,金英权,u23,于大宝
+6552402830475395342_!_103_!_news_sports_!_瓦基弗银行3比2击败科内利亚诺杀进欧冠决赛，朱婷受伤仍砍24分，你如何评价？_!_
+6552268722273255684_!_103_!_news_sports_!_季后赛如此表现的保罗值得火箭续约吗？_!_
+6552129155754885640_!_103_!_news_sports_!_曾以为他堕落平庸了，其实一直在蕴藏中，这就是我们最爱的隆多！_!_隆多,霍乐迪,季后赛,半决赛
+6552475433709863427_!_103_!_news_sports_!_《赏月夜·瞻世界》（第3期）-潘帕斯雄鹰盘旋天际只为最后一击？_!_肯佩斯,世界杯,原创专栏,潘帕斯雄鹰,阿根廷,马拉多纳
+6552286403084943619_!_103_!_news_sports_!_如何看待恒大0:3输给副班长大连一方，是故意的吗？_!_
+6552311549606756871_!_103_!_news_sports_!_膜拜！詹姆斯绝杀，全联盟众球员发推：历史最佳_!_詹姆斯,德拉季奇,绝杀,勒布朗,麦科勒姆
+6552455826961858824_!_104_!_news_finance_!_信用卡如何把临时额度提升到永久额度？_!_
+6552350716080423181_!_104_!_news_finance_!_如何5分钟获取3W元以上额度的信用卡技巧？_!_
+6552408668271280654_!_104_!_news_finance_!_做中国巴菲特系列之 涨停板环境研究（私募内幕）_!_系统环境,系统风险,涨停板战法,大盘,平衡市
+6552362706534400519_!_104_!_news_finance_!_A股纳入MSCI倒计时 利好四大板块_!_中国A股,大消费,每日额度,MSCI,A股,增量资金,国金证券
+6552342282748559630_!_104_!_news_finance_!_一斤两块五的青菜人都买的吃，为什么两块五的股票没人买呢？_!_
+6552248676465508877_!_104_!_news_finance_!_刚刚！马云正式宣布开放余额宝！4.7亿人喜极而泣！_!_开放余额宝,华为手机,支付宝,马云,天弘基金
+6552417212202746371_!_104_!_news_finance_!_碧生源(00926)拟溢价9.89%发行2.93亿股予Aurum Credo募资1.46亿港元_!_港元,碧生源,Aurum
+6552433853699457550_!_104_!_news_finance_!_2018千人小程序峰会_!_阿拉丁,新媒体
+6552338568788836622_!_104_!_news_finance_!_假如你的资产超过1000万，还需要出来工作或者上班吗？怎么安排比较好？_!_
+6552387771669938696_!_104_!_news_finance_!_农业银行日照岚山支行举办青年演讲比赛_!_五四,人民网,陈明强,农业银行
+6552165779234619661_!_104_!_news_finance_!_无抵押贷款都有哪些途径？_!_
+6552375660415811853_!_104_!_news_finance_!_没有雄厚资金的创业者，如何最大程度提高成功创业几率？_!_
+6552241668622385672_!_104_!_news_finance_!_惊恐“套路贷”是如何毁掉一个人的？_!_放贷人,借款,借款人,低条,套路贷
+6552364740272718093_!_104_!_news_finance_!_“老干妈”陶华碧恪守“坚决不上市”，你怎么看？_!_
+6552427528097628430_!_104_!_news_finance_!_办理下来的工商银行信用卡刚额度有多少？_!_
+6552157883839021316_!_104_!_news_finance_!_怎么来寻找和选择合适的投资商？_!_
+6552328653764559112_!_104_!_news_finance_!_如何对大脑进行投资该投资什么？_!_
+6552374765447807496_!_104_!_news_finance_!_这篇关于马云的文章，受益匪浅，我都看了很多遍了_!_潜意识,成功者,失败者,放松管制,创业机会
+6552401575493501447_!_106_!_news_house_!_南京嘉恒有山二期别墅住宅什么时候交付？南京嘉恒有山-云有山苑_!_金牛湖,度假区,400-060-1618转33361,南京,六合区
+6552278126360002824_!_106_!_news_house_!_昆明做房地产销售的前途可以吗？_!_
+6552408867140010253_!_106_!_news_house_!_在海口生活，是先买车还是先买房？_!_
+6552401230495219981_!_106_!_news_house_!_为什么青岛的房价还这么贵？_!_
+6552350873329074436_!_106_!_news_house_!_二手房中介法人代表卷款跑了怎么办？_!_
+6552389643214520590_!_106_!_news_house_!_开发商为什么急着把房子卖出去，而不是等到升值了再卖呢？_!_
+6552470677616591368_!_106_!_news_house_!_龙华新区观澜11栋带花园《福星家园》均价9800单间23万起可分期_!_福星家园,实用率,龙华新区,维也纳,精装修
+6552481361024778766_!_106_!_news_house_!_嘉陵又一商业中心即将崛起，坐拥都尉路城市中轴核心区！_!_乐汇城,广安市绿岛房地产开发有限公司,嘉陵区,嘉陵,BINGO
+6552478771436323332_!_106_!_news_house_!_“限酒令”后，杭州酒店式公寓冲上4万+！别高兴太早……_!_成交量,二手房,二手房价格,新房,交易量
+6552421303712743684_!_106_!_news_house_!_五月份是买房的好时间吗？_!_
+6552289099363909891_!_106_!_news_house_!_房价这么高，你觉得还有拼命干的必要吗？_!_
+6552303923607109896_!_106_!_news_house_!_杭州的房价走势如何？_!_
+6552290881183941124_!_106_!_news_house_!_小心“伪”学区房的陷阱_!_文字游戏,学区房,买房,购房者,楼盘
+6552453682326143236_!_106_!_news_house_!_2018年随着二胎政策放开，三四线城市房价还在上涨，这两者有何关系，房价上涨的原因有哪些？_!_
+6552375894151791111_!_106_!_news_house_!_调控严厉归严厉，土地买卖还是“马照跑、舞照跳”_!_张大伟,土地市场,一代宗师,二线城市,房地产企业,房企
+6552315966330503438_!_106_!_news_house_!_房产证上没有男方名字，男方的私生子能继承房产吗？_!_
+6549783743878398215_!_106_!_news_house_!_有人说买房贷款越多越好、还款时间越久越好，真的是这样吗？_!_
+6552392814460666115_!_106_!_news_house_!_西安房价均价降了11块，还有继续降的可能吗？_!_
+6552485295562949128_!_107_!_news_car_!_10kw风冷柴油发电机/大泽动力双缸柴油发电机_!_柴油发电机10kw风冷柴油,柴油发电机,双缸,风冷,大泽动力双缸柴油
+6552324378422411784_!_107_!_news_car_!_夏天汽车热得像“蒸笼”，太难受！聪明老司机这样做，30秒降温_!_降温剂,车载风扇,瞬间清凉干冰喷雾,暴风级,防滑垫
+6552409129191735556_!_107_!_news_car_!_想买7-9W自动挡二手车，有什么推荐？_!_
+6552362017762574596_!_107_!_news_car_!_20万上下SUV的，哪款比较好？_!_
+6552272705415021064_!_107_!_news_car_!_女司机开夜车跑长途，只带充气床还不够，后备箱带这些车品更安全_!_充气床,真空胎,女司机,后备箱
+6552295652259266830_!_107_!_news_car_!_为什么大众的车倒车时，后尾箱的把手（就是那个大众标）会翻起来？_!_
+6552397414077039112_!_107_!_news_car_!_CR-V直接放弃！本田祭出高颜值小型SUV配四驱，直接抢宝骏510蛋糕_!_颜值,摩托车,本田Vision,宝骏510,SUV
+6552283881565848071_!_107_!_news_car_!_「热点」高速上女司机突然的一个举动，一条命没了！4成事故皆因此发生！_!_变道,高速交警,匝道口实线变,交管局,匝道口
+6552341155793600775_!_107_!_news_car_!_同为17款手动的瑞虎5x豪华版和长安cs55炫色型，选哪个？_!_
+6552283261601579277_!_107_!_news_car_!_想买个2万元以下的二手车，不要面包车，哪个牌子比较好？_!_
+6552325303673291278_!_107_!_news_car_!_离奇！男子叫网约车，竟然叫来一辆……他回想起来还有点怕_!_车平台,网约车,顺风车,城事特搜,应急灯
+6552343565689684493_!_107_!_news_car_!_每天一款复古车｜70年老哈雷改装欣赏_!_
+6552256626609357320_!_107_!_news_car_!_十几年老摩托更换套缸，重新焕发青春！换缸过程全图记录_!_摩托车,机油,活塞销,燃烧室,活塞环
+6552360744493515267_!_107_!_news_car_!_超跑不输郭富城，有16块“8882”车牌，今年赚36亿，出门爱吃火锅_!_超跑,埃尔法,LP-700,V12,法拉利488,郭富城,沈嘉伟,兰博基尼
+6552143541273887245_!_107_!_news_car_!_豪车体验师：你这辆车与豪车之间的距离，只差了几个高科技车品_!_车载空气净化器,MP4,黑科技,HUD,洗车器
+6552397807917990414_!_107_!_news_car_!_成都雨泉新能汽车展销中心怎么样？媒体曝光后竟是这样……_!_展销中心,能源汽车,新能源汽车,新能源,雨泉新能源汽车
+6552447713378566413_!_107_!_news_car_!_四冲程摩托车发动机可以用来做油直吗？_!_
+6552402766755528973_!_107_!_news_car_!_30岁的男人应该开什么车？_!_
+6551980557121618436_!_107_!_news_car_!_客机成本那么高，为什么飞来飞去还能赚钱？今天总算知道了！_!_刹车组件,莱特兄弟,客机,刹车系统,320系列
+6552295967759008269_!_108_!_news_edu_!_高考倒计时开始！看各国考生花样搞事情，让压力变浮云_!_甩头照,泰晤士河,奥斯汀,兰开夏,德克萨斯大学
+6552377967555641859_!_108_!_news_edu_!_课程回顾｜中国古代建筑技术_!_古代建筑,中国古代,四合院,和玺彩画,旋子彩画
+6552425920076972292_!_108_!_news_edu_!_苏州大学全国排名第13位不太可能吗？_!_
+6552486920956412423_!_108_!_news_edu_!_从毕业到月薪过万，你用了多久？_!_Java,互联网,电子工程,月薪
+6551970219986780423_!_108_!_news_edu_!_如何看待人物《奥数天才坠落之后》一文以及付云皓的回应？_!_
+6552317037010485517_!_108_!_news_edu_!_高考专业相同或者相近，院校如何选择？_!_
+6552358612310688003_!_108_!_news_edu_!_有没有文凭很低，但知识水平很高的人？_!_
+6552431257089736973_!_108_!_news_edu_!_香港浸会大学的国际新闻认可度在香港和大陆到底如何？_!_
+6552463644439872008_!_108_!_news_edu_!_西北工业大学情系城固山区教育_!_苟秉宸,城固县,西北工业大学,水磑中心学校,镜观神十一,甲骨文
+6552417360274260494_!_108_!_news_edu_!_你真的会因材施教么？_!_职业规划,个性化,起跑线,心理学,大数据
+6552396930259878147_!_108_!_news_edu_!_《愚公移山》应不应该退出初中语文课本？_!_
+6552317416796324103_!_108_!_news_edu_!_大学生毕业在即，准备二战考研，但是学校逼着签三方协议，这样合适吗？_!_
+6552481107827228936_!_108_!_news_edu_!_为什么山东盛产学霸？_!_
+6552409193008070919_!_108_!_news_edu_!_眼睛近视想考军校，是高考前或是高考后做激光手术比较好？_!_
+6552168084675756296_!_108_!_news_edu_!_你每天读书几小时？_!_
+6552356268953043207_!_108_!_news_edu_!_考试，县级事业单位和公务员，应该选哪个？_!_
+6552338521464504845_!_108_!_news_edu_!_CAD基础级练习题使用构造线命令及详细答案（13）_!_三角形,构造线,垂直线,ABC,正五边形
+6552488928375472653_!_108_!_news_edu_!_两套一年级数学第五六单元测验卷，比你想的难，阶段测试尤为重要_!_年级
+6552475234245542148_!_109_!_news_tech_!_如果安卓系统手机没有流氓软件，安全性提高，国产机的份额是不是还会上升？_!_
+6552454631434223879_!_109_!_news_tech_!_迷你世界移动墙怎么做？_!_
+6552390951002702083_!_109_!_news_tech_!_电信的业务都有什么坑？_!_
+6552432983775642126_!_109_!_news_tech_!_太狠了！已有城市执行！有借呗、白条记录全都不能贷款_!_征信,借呗,信用卡,银行,消费贷
+6552258634011640077_!_109_!_news_tech_!_几乎每一个银行工作人员都劝说顾客，不要把钱存入余额宝，对此你怎么看？_!_
+6552157364059898115_!_109_!_news_tech_!_中国为什么不鼓励混合动力汽车？_!_
+6552253624167170308_!_109_!_news_tech_!_以后会有可能用无人机代替现在民航机吗？_!_
+6552374526905156099_!_109_!_news_tech_!_马云：阿里不愿招北大清华毕业生，他们应该到中小企业_!_内塔尼亚胡,阿里巴巴,马云,人工智能,科技创新
+6552376456519877123_!_109_!_news_tech_!_网易商丘诚寻合作伙伴 期待与您合作共赢！_!_雷哥定亲,网易邮箱,商丘,傻娘,网易商丘,微电影,信义兄弟,新媒体
+6552452578469216520_!_109_!_news_tech_!_拉卡拉、瑞和宝、随行付、立刷，大家推荐哪一款？_!_
+6552406036425212174_!_109_!_news_tech_!_逼死实体店的电商，反过来大量开实体店是为啥？_!_
+6552453444752376068_!_109_!_news_tech_!_为什么芯片那么难制造？_!_
+6552264093879763208_!_109_!_news_tech_!_就品质和质量而言，是小米好，还是OV好？_!_
+6552259057397268744_!_109_!_news_tech_!_为什么说“不惜一切代价发展芯片产业”是危险的？_!_
+6552340114595381511_!_109_!_news_tech_!_为什么在淘宝货源里有的可以分销有的不可以？_!_
+6552450239511396616_!_109_!_news_tech_!_如果百度的人工智能失败了，那么百度将何去何从？_!_
+6552462303411831043_!_109_!_news_tech_!_如何看待阿里钉钉CEO的言论：不知道你们10点之前下班干嘛，不到1点不合格？_!_
+6552180608288686343_!_109_!_news_tech_!_有什么高逼格的微信名字？_!_
+6552319886498988302_!_109_!_news_tech_!_经常收到银行的办信用卡的信息，说是白金优质用户，是真的吗？_!_
+6552275721006350596_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552350518465790215_!_110_!_news_military_!_叙利亚的前途在哪里？_!_
+6552449382967410958_!_110_!_news_military_!_次声武器属于第几代武器，它发展到什么程度了呢？_!_
+6552287273713402371_!_110_!_news_military_!_美军新型战略杀器即将问世，我军轰20还未研制完成！_!_B-1B,B-2A,轰炸机,美国空军,电子战,B-21,轰-6,B-2
+6552298036813365518_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552147754712302093_!_110_!_news_military_!_五角大楼清醒认识到一个现实：在这些领域中国正在迎头赶上！_!_五角大楼,人工智能,大气层,无人机,战斗机
+6552120538314048007_!_110_!_news_military_!_苏-57战机红场首秀，俄空军还让刚公布的杀手锏武器参加阅兵_!_战机,武器,阅兵,苏-57,俄罗斯
+6552294787846766851_!_110_!_news_military_!_美国为何有底气在世界上挑起战争？_!_
+6552277802022863374_!_110_!_news_military_!_美国海军重建第二舰队 遏制俄罗斯海上力量扩张_!_北大西洋,北大西洋公约组织,舰队,2018美国国防战略报告,海军,俄罗斯
+6552498221485130243_!_110_!_news_military_!_俄罗斯主战装备防空导弹被掀底牌，全靠一邻国“热心”卖给美军_!_防空导弹系统,S300,俄罗斯,乌克兰,苏27
+6552289611446485252_!_110_!_news_military_!_为什么经济已经十名开外的俄罗斯，还能够稳居世界军事强国，并和美国针锋相对？_!_
+6551317889725497614_!_110_!_news_military_!_日本的发动机为什么那么先进？特别是摩托车的发动机？_!_
+6552379602528895236_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552140698911506948_!_110_!_news_military_!_美国天天打仗，为什么还那么有钱_!_保护费,华盛顿海军条约,美国,薅羊毛,盟友,打战
+6552262364899574024_!_110_!_news_military_!_中国辽宁航母战斗群的战斗力如何？_!_
+6552464775014515214_!_110_!_news_military_!_美或借“协议”出手遏制伊朗_!_叙利亚,特朗普,伊朗,以色列,伊核协议
+6552149648201482499_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552383331122545160_!_112_!_news_travel_!_梦幻西游：让时光倒流，重回到2005年，看一下那时的梦幻攻略_!_梦幻西游,时光倒流
+6552282823670104589_!_112_!_news_travel_!_世界上原始生态环境保存较好的自然保护区-可可西里_!_
+6552308149427110151_!_112_!_news_travel_!_2个人去西安旅游，从长春出发，大概要多少钱？_!_
+6552312710644630029_!_112_!_news_travel_!_一处处秀丽的自然风光让您目不暇接 没去过的赶紧去_!_大草原,杜鹃林,骊山,兵谏亭,长海子
+6552468264906129924_!_112_!_news_travel_!_珠海这个绝美的海景驿站升级改造完成！空气清新，骑行晨跑便利！_!_海景,凤凰湾,龙船花,情侣路,驿站
+6552288619682333198_!_112_!_news_travel_!_三清山，世界上已知花岗岩地貌中分布最密集、形态最多样的峰林_!_
+6552469102479278343_!_112_!_news_travel_!_四川游客张某带着六名女子在三亚沙滩烧烤被撒沙，这事你怎么看？_!_
+6552303954431050243_!_112_!_news_travel_!_为何江西能成为国内民众五一假期游的首选之地_!_明月山,婺源,太平山,大觉山,江西
+6552283572516946183_!_112_!_news_travel_!_美人美景美时节，下一句怎么接？_!_
+6552365404885352973_!_112_!_news_travel_!_重回北川老县城 参观者更懂得珍惜当下_!_唐家山堰塞湖,北川老县城地震遗址,汶川特大地震,景科凤,讲解员
+6552401005902823944_!_112_!_news_travel_!_10元人民币，斯里兰卡一顿饭吃不完，中国一顿饭吃不饱_!_牛油果,斯里兰卡,印度洋,Galle,斯里兰卡卢比
+6552246456852414980_!_112_!_news_travel_!_山村50多年前制作的粮食加工机械，现在还能使用，城里人都没见过_!_
+6552483174889292040_!_112_!_news_travel_!_国内哪一家航空公司的飞机餐最好吃？_!_
+6552186008069210631_!_112_!_news_travel_!_赴英旅游正当时 盘点英国游景点_!_远离尘嚣,斯凯岛,苏格兰,国家地理杂志,英国,撒克逊,白金汉宫
+6552292924552380941_!_112_!_news_travel_!_夏威夷四岛攻略｜这样玩才叫去过Aloha！_!_可爱岛,茂宜岛,夏威夷,威基基,夏威夷州
+6552289068397363716_!_112_!_news_travel_!_深圳大鹏葵坝老路变身山水绿道_!_深圳,大鹏半岛,徒步爱好者,葵坝老路,排牙山
+6552290498453701128_!_112_!_news_travel_!_夏日拍花儿好去处：滑县森林公园里的“月季园”芳香四溢味正浓_!_月季园,五一,滑县森林公园设计规划,前夕正式对外开放,游客观赏休憩地
+6552302167980507655_!_112_!_news_travel_!_美国夏威夷火山爆发，同时伴随6.9级地震，一千度岩浆溢出_!_夏威夷,基拉韦厄,岩浆,火山活动,火山爆发
+6552347122748883208_!_112_!_news_travel_!_G7京新高速从天津出发到新疆沿途怎么玩好？_!_
+6552483297979531784_!_113_!_news_world_!_印度的这些习俗中，除了左手上厕所外，还有这个习俗我理解不了_!_恒河水,印度人,印度,恒河沐浴,习俗
+6552151374707032583_!_113_!_news_world_!_亚洲最善战的国家，四周全是敌国，天天打仗，至今比日韩还富裕！_!_黎巴嫩,犹太人,以色列,约旦,叙利亚
+6552473688036344333_!_113_!_news_world_!_不知道的秘密：火车退役后都去了这些地方，实在是令人毛骨悚然_!_北卡罗来纳州,火车墓地,Joo,毛骨悚然,玻利维亚
+6552466773604565255_!_113_!_news_world_!_日本人如何夺走中国烹饪的世界地位？_!_
+6552260640910606605_!_113_!_news_world_!_有哪些科学解释不了的事实或现象？_!_
+6552281708425642504_!_113_!_news_world_!_特朗普否认给艳星13万美金封口费 不料律师说漏了_!_律师,艳星,白宫发言人,特朗普,封口费
+6552166158735245575_!_113_!_news_world_!_美国建国不过两百年，为什么会成为世界强国？_!_
+6552435352982782212_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552154143354519815_!_113_!_news_world_!_特朗普为何称美国不想再当世界警察，只想管好自己？_!_
+6552261172517667335_!_113_!_news_world_!_伊朗风俗禁忌_!_土库曼斯坦,蔷薇园,医典,德黑兰,土库曼,列王记,阿塞拜疆,伊朗
+6552430873851986439_!_113_!_news_world_!_这个亚洲最土豪的国家，民众生活政府掏钱，但却为娶媳妇而发愁_!_文莱人,文莱,南朝鲜,沙巴州,砂拉越州
+6552396655189033480_!_113_!_news_world_!_普京视察时橘猫“小桥”抢镜，背后故事暖心_!_小桥,普京
+6552404301883375879_!_113_!_news_world_!_美国527委员会究竟是什么，有什么作用？_!_
+6552379341014040835_!_113_!_news_world_!_为什么罗斯福新政时宁愿把过剩的牛奶、棉花扔掉也不免费分发给穷人？_!_
+6552286024809054478_!_113_!_news_world_!_俄罗斯在普京之后是否有可能再次发生解体？_!_
+6552306166368567812_!_113_!_news_world_!_一箭双雕！美民众点评特朗普“挺拥枪”：虚伪_!_达拉斯市,校园枪击,得克萨斯州,共和党,枪支
+6552135329598406926_!_113_!_news_world_!_日本地铁为什么没有安检？_!_
+6552222614033531396_!_113_!_news_world_!_已成为美俄刀俎上的鱼肉，乌克兰的出路到底在哪里？_!_原苏联,乌东,俄罗斯,乌克兰,波罗申科
+6552270626663432456_!_113_!_news_world_!_普京在刚上台时说要带领俄罗斯成为强国，就目前的形势来看，俄罗斯是否成为了强国？_!_
+6551238301393617416_!_115_!_news_agriculture_!_玉米到底用不用追肥？怎样做更高产？专家为你解惑_!_玉米,种植户,玉米生长,追肥,螯合锌
+6552364912784441859_!_115_!_news_agriculture_!_这种花太好养活了！2个月长满阳台，庭院变花海，拦都拦不住！_!_点击图片选购,爬藤三角梅,三角梅,灿若云霞
+6552385983290016269_!_115_!_news_agriculture_!_河南叶县扶贫广播“打通扶贫政策落地最后一公里”_!_我们一起奔小康,扶贫,艾叶草,杜金龙,精准扶贫，晨钟送暖,扶贫政策,叶县
+6552285591235461645_!_115_!_news_agriculture_!_午餐要吃好，荤素搭配不能少，这菜我看着很喜欢！_!_午餐,营养
+6552429612490555917_!_115_!_news_agriculture_!_跟天赌与地斗，如今行情摸不透，老菜农一句话总结56年种菜辛酸_!_张大叔,曲曲菜,野菜,荠菜,种菜
+6552394464827015687_!_115_!_news_agriculture_!_子女农转非后，承包地将不允许继承，满足6个条件，可以非转农！_!_农业户口,申请非转农,承包地,农转非,农村
+6552308600583225869_!_115_!_news_agriculture_!_青岛乡医人均收入增加近万元 村卫生室药物补助提高_!_村卫生室,老年乡村医生生活补助,高等专科学校,医学生,定向培养
+6552358500494737928_!_115_!_news_agriculture_!_梨树上结“金子”，树下生“票子”，来看看他的生财之道_!_黄金梨,杀虫灯,土鸡,正阳县,李军
+6552371596881822211_!_115_!_news_agriculture_!_一农村妇女搞养殖年入百万 能学到她的一半就能赚大钱_!_
+6552304438910910983_!_115_!_news_agriculture_!_酒糟新用途？酒糟分离机助力！_!_赖氨酸,啤酒糟,酵母菌,分离机,白酒糟
+6552294439178469891_!_115_!_news_agriculture_!_两岸不交流，大陆停止收购台南凤梨，果农欲哭无泪！_!_凤梨园,台南,台湾凤梨,种植业,金钻凤梨
+6552476434709873160_!_115_!_news_agriculture_!_注意了，农村今年有4个坑，有农民已经愁坏了_!_河南省,中牟县,农转非,非转农通道,农村
+6552262937292046852_!_115_!_news_agriculture_!_80后农民种菜致富1年纯赚70万 他发明1个销售绝招让他快速致富_!_谢锋,摩托车,无公害蔬菜,绿色蔬菜,蔬菜大棚
+6552449400507990276_!_115_!_news_agriculture_!_哪种发酵剂发酵猪饲料比较好？_!_
+6552484878288421390_!_115_!_news_agriculture_!_怎样购买药材种子_!_中药材,用种量,出苗率,坏种子,千粒重
+6552360588352160270_!_115_!_news_agriculture_!_文化周刊｜中国传统村落保护（海南）高峰论坛儋州举行_!_传统村落保护,张岳崧,儋州,高林村,海南
+6552378152583168270_!_115_!_news_agriculture_!_做好基层农村工作需要哪些基础知识？_!_
+6552260269102334215_!_116_!_news_game_!_刺激战场中比较难获得的称号有哪些？_!_
+6552381422257045773_!_116_!_news_game_!_《坦克世界》里的KV85怎么玩？_!_
+6552337235113410823_!_116_!_news_game_!_你玩过剧情最好的RPG游戏是什么？_!_
+6552348565958557966_!_116_!_news_game_!_王者荣耀各位置开局应如何处理更为合理？_!_
+6552411854570783239_!_116_!_news_game_!_英雄联盟里的十个经典笑话：为何放弃治疗，因为我要带闪现！_!_英雄联盟,三十一,游戏,女朋友,锤石
+6552425352965128451_!_116_!_news_game_!_跑跑卡丁车为何被QQ飞车击败？_!_
+6552327759589277955_!_116_!_news_game_!_如果《我的世界》和《迷你世界》动用所有力量“开战”会怎样？_!_
+6552281885177807368_!_116_!_news_game_!_DNF：六人raid2保4!第二脊椎了解一下！_!_罗特斯,安徒恩,GBL,阿甘左,邪光斩
+6552034677706719757_!_116_!_news_game_!_拿了98K因枪法太烂，一枪都打不中？教你称霸绝地求生的独门绝学_!_枪法,战术大师,自动步枪,敌人,射击
+6552276567731470596_!_116_!_news_game_!_DNF武器加13成功是一种什么样的体验？_!_
+6552326359681597703_!_116_!_news_game_!_LOL哪个区玩家最有素质？_!_
+6552468899986669832_!_116_!_news_game_!_王者荣耀里坦克出不出霸者呢？_!_
+6552329910159933965_!_116_!_news_game_!_英雄联盟：UZI直播时间确定，粉丝疯狂舔屏！UZI女朋友都害羞了_!_uzi,英雄联盟,无冕之王,直播间,UZI
+6552455768606507523_!_116_!_news_game_!_「学技术」eStarPro.星辰裴擒虎爆发出装铭文推荐_!_收割型,爆发型,物理英雄,有助于,诸葛亮
+6552381885241098759_!_116_!_news_game_!_梦幻西游：土豪被美女骗钱骂屌丝，一怒之下买175张号复仇_!_梦幻西游,游戏
+6552446187478188551_!_116_!_news_game_!_动漫cosplay《王者荣耀》牛一样的眼睛浓妆甄姬—游园惊梦_!_cosplay,眼睛浓妆甄姬—,王者荣耀,游园惊梦,甄姬—游园惊梦动漫cosplay,甄姬
+6552107641571639822_!_116_!_news_game_!_阴阳师：这几位SSR基本全靠碎片合成 不是怕脸黑而是实用度太高了_!_式神,辉夜姬,碎片,彼岸花,SSR
+6552482832072049160_!_116_!_news_game_!_iBoy直播挂机惨遭LPL官方罚款一万元，打野螳螂或将面临封号？_!_挂机,iBoy,英雄联盟,螳螂,EDG
+6552433522764677384_!_101_!_news_culture_!_上联：雨后美景天际来，下联怎么对？_!_
+6551017661524869383_!_101_!_news_culture_!_为什么有那么多人喜欢林冲？_!_
+6552131686069436679_!_101_!_news_culture_!_同样是喜剧大师，为什么在很多观众心中陈佩斯是艺术家，而赵本山不是？_!_
+6552237434510246148_!_101_!_news_culture_!_哪些网络小说值得推荐？_!_
+6552286502829687053_!_101_!_news_culture_!_上联，一石击激起千层浪，请接下联？_!_
+6552433692315222286_!_101_!_news_culture_!_如何评价硬笔书协主席张华庆的书法水准？_!_
+6552140591189197064_!_101_!_news_culture_!_上联：春江花开月明夜，下联怎么对？_!_
+6552366881993392392_!_101_!_news_culture_!_上联：本欲君临天下，下联怎么对？_!_
+6552276740436132110_!_101_!_news_culture_!_上联：屋后梅花三两亩；求下联怎么对？_!_
+6552366428748513795_!_101_!_news_culture_!_用扬琴讲一段故事……_!_林冲,弹词三六,倒垂帘,苏武,水浒,扬琴
+6552226834551931405_!_101_!_news_culture_!_《晨夏》笔起_!_晨夏
+6552289764379197955_!_101_!_news_culture_!_历史的见证，国人的楷模—原子城纪念馆_!_原子城,纪念馆,原子城纪念馆,原子弹,展厅
+6552287064895783181_!_101_!_news_culture_!_上句：麻将桌上观君子，如何接下句？_!_
+6552463835150680579_!_101_!_news_culture_!_道德经第十章_!_
+6552453592857444868_!_101_!_news_culture_!_中国古代“三无”状态，少妇如何相思相怨，一个“月”字足矣_!_闺怨诗,征妇怨,思妇,相思,春江花月夜,独不见,杂诗三首,自君之出矣
+6552406537237692936_!_101_!_news_culture_!_广州新观察：传统文化与城市精神_!_城市名片,博士生导师,岭南文化,文化,地理分布
+6552402160112369924_!_101_!_news_culture_!_旗袍是清朝的服饰吗？_!_
+6552277737296363789_!_101_!_news_culture_!_上联：五粮香天下，怎么对下联？_!_
+6552307612396814605_!_101_!_news_culture_!_上句：麻将桌上观君子，如何接下句？_!_
+6552367237188026628_!_102_!_news_entertainment_!_金粉世家要翻拍，刘诗诗演冷清秋，她和董洁谁更适合民国风？_!_
+6552439163889648131_!_102_!_news_entertainment_!_谢娜微博粉丝破亿，获两项吉尼斯世界纪录惹热议！网友发图打脸_!_吉尼斯世界纪录,快本,谢娜,双喜临门,好莱坞,僵尸粉
+6552447844714807815_!_102_!_news_entertainment_!_全球票房超过10亿美元的电影有哪些_!_爱丽丝梦游仙境,美女与野兽,玩具总动员3,变形金刚3,速度与激情,疯狂动物城,复仇者联盟,星球大战,黑暗骑士,海底总动员2,加勒比海盗,钢铁侠3,侏罗纪公园,冰雪奇缘,侏罗纪世界
+6552262935215866376_!_102_!_news_entertainment_!_谢娜强势复出后迎37岁生日，快乐家族凌晨齐送祝福_!_谢娜,李维嘉,张杰,吴昕,快乐家族
+6552296398690189827_!_102_!_news_entertainment_!_谢娜何炅讨论下辈子成为彼此的谁，两人答案看哭网友_!_生日快乐,谢娜,何炅,下辈子,土鸡蛋
+6552355063778509326_!_102_!_news_entertainment_!_郭珍霓真是胆大！上面看着好难受啊，网友：不是身材好根本挂不住_!_粉丝,郭珍霓
+6552412564033110280_!_102_!_news_entertainment_!_皇后因庶出身份被华妃嘲笑，皇后面不改色，为何甄嬛却怒了呢？_!_
+6552463998640456199_!_102_!_news_entertainment_!_YY女神豪jenny婷婷姐呼吁粉丝勿骂仔仔 他是个单纯的主播_!_仔仔,粉丝,女神,婷姐
+6552281226969874948_!_102_!_news_entertainment_!_杨颖出席活动，网友：一头精美的发型和她完美的气质_!_杨颖出席活动,一头精美
+6552367747165061639_!_102_!_news_entertainment_!_这部港剧未拍先火只因杨千嬅的加入，能否拿下年底TVB大奖？_!_杨千嬅,妙手仁心2,真情,天地真情,少女的祈祷,TVB,我本善良,金像奖,多功能老婆,再见亦是老婆,陈宝华,美味情缘,黄浩然
+6552351390142824974_!_102_!_news_entertainment_!_抖音网红费启鸣是怎么火起来的？减肥前180斤、不知名二本学院_!_粉丝,快乐大本营
+6552311982047887885_!_102_!_news_entertainment_!_电影《后来的我们》中的爱情金句你都能看懂的话，就算你赢_!_刘若英,爱情故事,电影,姚家镇,后来,爱情电影
+6552254473400484355_!_102_!_news_entertainment_!_最新一期快乐大本营，吴昕怀孕，奚梦瑶被黑惨了！_!_杨迪,快乐大本营,吴昕,奚梦瑶,蔡少芬
+6552224051798999565_!_102_!_news_entertainment_!_美国明星出席活动街拍，网友表示很有气质很养眼！_!_Isabell,东方IC,抵达Arvina街头街拍,Axl,Warburton
+6552829913357877763_!_102_!_news_entertainment_!_周扬青小三上位？女星Z买通稿抹黑对手？许魏洲造型太奇葩？_!_阎鹤祥,佘诗曼,李毓芬,清秀佳人,岳云鹏
+6552301280360923662_!_102_!_news_entertainment_!_张一山生日 杨紫为其画画 周冬雨P图调侃好友_!_周冬雨,放狠话,杨紫,张一山,法令纹
+6552417521864016388_!_102_!_news_entertainment_!_Netflix翻拍的《西游记》，人设颠覆，五毛特效！_!_纸牌屋,孙悟空,美国囧案,唐三藏,花游记,悟空传,大话西游,怪奇物语,救世主,国王的演讲,白夜追凶,Netflix,西游降魔篇,西游记,新猴王传奇
+6552360077796311304_!_102_!_news_entertainment_!_爱丽丝和杨家将能和平相处么？_!_
+6552319177858744836_!_102_!_news_entertainment_!_张靓颖演唱会伤心流泪，冯轲看比赛逍遥自在，微博发文让人心寒_!_张靓颖,冯轲,演唱会,尚雯婕,利物浦
+6552281003467997699_!_102_!_news_entertainment_!_48岁钟丽缇一袭红裙出席活动，网友：越活越年轻了！_!_钟丽缇一袭红裙出席活动,钟丽缇,红裙,越年轻
+6552491926686269960_!_103_!_news_sports_!_NBA季赛后，东西部半决赛局势不同，詹姆斯有望再破纪录_!_爵士队,詹姆斯,骑士队,季后赛,凯尔特人队
+6552254893132873992_!_103_!_news_sports_!_上海上港输球，山东鲁能有机会赢球超越上港排名中超第一吗？_!_
+6552403523198255624_!_103_!_news_sports_!_浙体产业周刊｜浙体产业一周资讯速递2018-010_!_合成材料面层健身步道要求,报告,马术,山地自行车,周刊,浙江省,马拉松,绝地求生
+6552392786564350468_!_103_!_news_sports_!_全国篮球青训看山东，济南体校成冠军摇篮_!_中国篮协,吴轲,省运会,朱荣振,青年队
+6552156835774726670_!_103_!_news_sports_!_《灌篮高手》全国大赛图文解说（四）_!_灌篮高手,安西教练,樱木,丰玉,宫城
+6552465743542223364_!_103_!_news_sports_!_中超第9轮最佳阵容：大连一方成大户 四大U23抢镜中超_!_广州恒大,大连一方,卡拉斯科,李帅,惠家康
+6552373007963128071_!_103_!_news_sports_!_世乒赛男团冠军即将产生，国乒男队能实现史无前例的九连冠霸业吗？_!_
+6552308145614487811_!_103_!_news_sports_!_足球比赛中守门员什么情况下可以用手，什么时候又不能？_!_
+6552331233869693444_!_103_!_news_sports_!_无惧流言再创记录，探花：我们的任务是尽快送他们去“见上帝！”_!_塔图姆,波士顿凯尔特人,莱昂纳德,费城,凯尔特人
+6552441606211895816_!_103_!_news_sports_!_2018-5-6 切尔西 vs 利物浦，今天继续一场稳胆英超冲击4连红_!_利物浦,英超,斯坦福桥,欧冠,切尔西
+6552326065786716424_!_103_!_news_sports_!_中国队亚洲杯抽到上上签，为何里皮脸色仍然很凝重？_!_
+6552794527541756419_!_103_!_news_sports_!_马内：利物浦需要费基尔，他来了将火力全开_!_费基尔,利物浦,里昂,奥拉斯,火力全开
+6552153218493710856_!_103_!_news_sports_!_马刺助教梅西纳将在今日面试雄鹿主帅职位_!_马刺,埃托雷,梅西纳,助教
+6552392777693397512_!_103_!_news_sports_!_老鱼绝杀：爵士陷困境无调整余地，主场战火箭也无优势_!_卢比奥,韦斯特,伊格达拉,米切尔,火箭队
+6552139867495596551_!_103_!_news_sports_!_连续第三次在世乒赛会师决赛，日本女队都只拿到银牌_!_
+6552134519980294413_!_103_!_news_sports_!_世乒赛女团决赛，世界第一快刘诗雯2:3爆冷输球。你怎么看？_!_
+6552270600147042567_!_103_!_news_sports_!_周琦为塔克唱生日歌，全队被逗的哈哈大笑，你怎么评价？_!_
+6552264805024006670_!_103_!_news_sports_!_连续5场20+，绿军20岁新秀超越伯德，如此数据不输米切尔西蒙斯_!_凯尔特人,西蒙斯,塔图姆,米切尔,季后赛
+6552387093954298372_!_103_!_news_sports_!_新疆王朝陨落？李秋平卸任主帅，3危机令新疆难争冠_!_新疆队,上海队,CBA,李秋平,常规赛
+6552293048561172744_!_104_!_news_finance_!_雪迪龙这家公司的股价为什么一直走低？_!_
+6552367244968460808_!_104_!_news_finance_!_玉米价格还会继续回落吗？_!_DDGS,双反,玉米,玉米拍卖
+6552378506800529924_!_104_!_news_finance_!_郑眼看盘：医药券商白酒为何脉冲式上涨？_!_权重股,MSCI,白马股,CDR,投资者,风水轮流转
+6552459010052718862_!_104_!_news_finance_!_“国六”即将实施，常用的“国三”“国四”会怎样处理呢？_!_
+6552490721402683661_!_104_!_news_finance_!_云联惠为什么会进博鳌论坛？_!_
+6552386629539987981_!_104_!_news_finance_!_2018年全国GDP增速排名说明了什么？您怎么看？_!_GDP增速,贵州,天津,第一季度,海南
+6552262995395740167_!_104_!_news_finance_!_这么多公链崛起，到底谁能走到最后？_!_阿萨姆邦,ZIL,ETH,Zcash,EBAY
+6552289463127507463_!_104_!_news_finance_!_我与一带一路沿线国家进口增速超出口_!_大数据,贸易合作,一带一路,贸易额,国家信息中心
+6552386144993018371_!_104_!_news_finance_!_上市近3月消失1900亿！股价疯狂下跌至41.8%！8万小散都傻眼了！_!_股民,股票,上交所,江南嘉捷,股东
+6552352970502045956_!_104_!_news_finance_!_面对危机餐饮行业该如何发展？_!_
+6552459966429528327_!_104_!_news_finance_!_被冷落后应该怎样调整自己的心态？_!_
+6552367070502191368_!_104_!_news_finance_!_负债累累，如何创业？_!_
+6552389769907667464_!_104_!_news_finance_!_人民币国际化：为“一带一路”保驾护航_!_人民币,西班牙,阿联酋,新加坡,支付手段
+6552155259769192712_!_104_!_news_finance_!_信用卡是这个月花下个月还，还没有手续费，那银行开信用卡业务是靠什么赚钱呢？_!_
+6552260265151299853_!_104_!_news_finance_!_快三十岁了找不到工作，也没资金创业，该何去何从？_!_
+6552298238165123598_!_104_!_news_finance_!_广东琶洲展，广交会阳江企业成交2.37亿美元_!_广交会,广东,阳江,采购商,本届广交会
+6552476439910809864_!_106_!_news_house_!_银行加息对房价走势有何影响？_!_
+6552282571118477572_!_106_!_news_house_!_私房拆迁分的房子是使用权，父母过世后，不在户口内的子女有继承权吗？_!_
+6552302863589048580_!_106_!_news_house_!_买了高层西边户会不会夏天很热，会受到什么影响？_!_
+6552490238256611588_!_106_!_news_house_!_如果房价突然降了，大家都买得起房子了，接下来会怎么样？_!_
+6552428244312785160_!_106_!_news_house_!_如果将要取消物业，大家会支持吗，物业到底该不该存在？_!_
+6552296073187033607_!_106_!_news_house_!_七栋功能齐全别墅，最后一栋就差一个好（壕）房主_!_别墅
+6552465468672704782_!_106_!_news_house_!_农村建房，明明砖混结构便宜，为什么有人还会要框架结构？_!_
+6552443890488574216_!_106_!_news_house_!_海南买房先签订购房合同，挂靠公司交两年社保，满两年后再过户，这样做靠谱吗？_!_
+6552372782565425415_!_106_!_news_house_!_海口和琼海，哪个地方更适合居住，房价大概多少？_!_
+6552387009254523140_!_106_!_news_house_!_未来房价会跌到什么程度？_!_
+6552404612844880387_!_106_!_news_house_!_5月6日孝感房产网签22套 成交均价6544元/㎡_!_香港城,毛陈镇,非住宅,孝感房产销售,房产网
+6552346485894152452_!_106_!_news_house_!_在青岛，夫妻俩每人一个月6000，是什么水平？_!_
+6552394732138397959_!_106_!_news_house_!_你对郑州尬舞团被取缔了怎么看？_!_
+6552353853725999619_!_106_!_news_house_!_如何用最少的钱在北京市中心买房_!_马德里,房价,房产,西班牙,买房
+6552221489477714440_!_106_!_news_house_!_万科4月合同销售额418.9亿元 新增22地产项目_!_万平方米,需支付权益价款,万科,建筑面积,合同销售
+6552216826950975751_!_106_!_news_house_!_现在一万一平米的房子，十年后大概卖多少钱？_!_
+6552265789846585613_!_106_!_news_house_!_中国楼市不断上涨，李嘉诚为什么加速撤离中国楼市？_!_
+6552254326922805507_!_106_!_news_house_!_想在西安买一套公寓，总价40万左右的有吗？_!_
+6552478142617878792_!_107_!_news_car_!_国产suv哪个品牌有底盘研发能力？_!_
+6552364702259741192_!_107_!_news_car_!_18款大众凯路威2.0T 商务豪车底价惠全国_!_凯路威,T6,电动门,大众凯路威,大众T6,天津市
+6552403536037020164_!_107_!_news_car_!_2018年第一季度车企销量增幅排行榜！看看哪个品牌势头最旺！_!_MG,广汽本田,MG6,荣威RX5,吉利汽车,荣威,锐腾,RX3,RX5,上汽,自主品牌,RX8
+6552244411332296974_!_107_!_news_car_!_预算50万左右，雷克萨斯ES、宝马五系、奔驰E级、奥迪A6这几款车应该怎么选择？_!_
+6552443169722597892_!_107_!_news_car_!_赛车手的倾心之作！一不小心用了晓乐同款_!_漫步者,赛车,何晓乐,纽博格林,赛车手
+6552435716331143428_!_107_!_news_car_!_金茶花该怎样养护？_!_
+6552257892035068424_!_107_!_news_car_!_别再让车子暴晒了，今年超火全自动车篷震撼上市，易收易放贼好用_!_汽车音响,USB车充,全自动,挡泥板,MP3
+6552430605315866893_!_107_!_news_car_!_预算20万之内，有什么适合家用以及上下班使用的汽车推荐？_!_
+6552392219565752589_!_107_!_news_car_!_观致5的真实油耗是多少？_!_
+6552320913252024590_!_107_!_news_car_!_夏季汽车空调该如何保养？_!_
+6552313459655049741_!_107_!_news_car_!_这才是老百姓能接受的良心车！比蔚领便宜5万仅6万，还买啥宝骏_!_骏派CX65,旅行车,手动变速箱,天津一汽,宝骏
+6552261920097829123_!_107_!_news_car_!_博瑞1.8T尊贵型和传祺GA8尊享版，选择哪款比较好？_!_
+6552490565663982094_!_107_!_news_car_!_世界上最贵的手动挡轿车，没有安全气囊ESP，160万比奔驰S级还贵_!_奔驰S级,保时捷911,手动变速箱,老爷车,手动挡,摩根Roadster,Roadster
+6552329616617373966_!_107_!_news_car_!_2018力狮eyesight版本的脚下为什么是三个踏板，这个三个踏板都是什么设置？_!_
+6552316648659878403_!_107_!_news_car_!_越来越多人不用GPS了，现流行民用“北斗”导航，高级倍享受_!_逆变器,防碰瓷,GPS,记录仪,密封胶
+6552253258209952260_!_107_!_news_car_!_别再傻傻给车贴膜了，聪明司机都这样防晒，比开空调凉爽一万倍_!_无骨雨刷,黑科技,雨刮片,反射镜,防眩镜
+6552379341894844941_!_107_!_news_car_!_即日起，等红绿灯玩手机也要扣分？这些你不知道的扣分行为！_!_红绿灯,电子狗,安全带,交通法规,红灯时
+6552238441055126020_!_107_!_news_car_!_18万入手这辆国产豪华SUV，开了一万五公里这油耗你能接受吗_!_SUV,WEY VV5,国产车,比亚迪,德系车
+6552148400035332621_!_107_!_news_car_!_人民币成该国法定货币，中国免其2亿欠款，奇瑞成当地豪车_!_摩托车,奇瑞,津巴布韦,通货膨胀,国产汽车
+6552298295585145352_!_107_!_news_car_!_意外遭他人车祸车辆撞击 Google自驾车再成事故车_!_钱德勒,山景城,Google自驾车,自驾车,自动驾驶
+6552367604269318407_!_108_!_news_edu_!_高考只剩一个月，应该怎样复习？_!_
+6552401551871181326_!_108_!_news_edu_!_学习马克思斗争精神：主义真，同志以奋斗不止_!_马克思,共产党宣言,恩格斯,马克思主义,斗争,物质力量
+6552309779404947715_!_108_!_news_edu_!_流浪狗扰乱正常教学秩序，破坏学生们生活，该怎么办？_!_
+6552294692162109703_!_108_!_news_edu_!_现在学什么专业好？_!_
+6552351715587260676_!_108_!_news_edu_!_辽宁男孩考上大连某高校，父母办升学宴庆祝幅超过千米，网友批评太奢侈，你怎么看？_!_
+6552382773309473293_!_108_!_news_edu_!_泰山医学院同等学力申硕火热报名中……_!_研究生,山东医学院,泰山医学院,学位证书,齐鲁大学
+6552316707547906574_!_108_!_news_edu_!_哈佛商业评论：如何让自己与令自己厌烦的人感同身受_!_It,McKee,Gloria,Instead,Fernandez,What,Louis,says.
+6552262819427910148_!_108_!_news_edu_!_学高为师推荐：小学一年级下册语文句子题（附答案）_!_跳绳,小猫,大火球,真可爱,大熊猫
+6552268579801137678_!_108_!_news_edu_!_唐山今早必知的那些事儿_!_义务教育,深圳市,学位补贴,教育券,唐山
+6552327110206161165_!_108_!_news_edu_!_为什么江苏的考生总说他们有多牛逼？_!_
+6552432142486339843_!_108_!_news_edu_!_中专学电脑编程好不好？_!_
+6552401937512268302_!_108_!_news_edu_!_「教资结构化」一个小孩说方言，其他学生不喜欢他怎么办？_!_教师,说方言,说普通话,普通话
+6552384099615506702_!_108_!_news_edu_!_你认识的高考严重超常发挥考入名校的人后来怎么样了？_!_
+6552428576921092616_!_108_!_news_edu_!_5个基本原则，让孩子学习英语事半功倍_!_英语教育,英语学习,关键期
+6552379392352322062_!_108_!_news_edu_!_鹿邑二高纪念“五四运动”99周年暨“一红两优”表彰大会_!_刘宏彬,五四,共青团员,映象网
+6552489225269281028_!_108_!_news_edu_!_学俄语有什么好的语法书推荐？_!_
+6552129138570821896_!_108_!_news_edu_!_为什么你的阅读速度慢，理解能力低，存在什么问题导致这样的现象？_!_
+6552464751203451400_!_109_!_news_tech_!_SSG甜蜜购：区块链+电商 新零售2.0时代杨帆起航_!_技术优势,区块链,区块链技术,SSG甜蜜购,新零售
+6552461872694558984_!_109_!_news_tech_!_微信支付是否已远远超过支付宝？_!_
+6552457750721331716_!_109_!_news_tech_!_离开高通小米将会怎样？_!_高通,核心技术,全球化,小米手机,技术能力
+6551711832342528269_!_109_!_news_tech_!_大多APP在评论区都不支持图片格式的评论内容，是不够刚需还是过于低频？_!_
+6552410841138856200_!_109_!_news_tech_!_你如何看待个人信息严重泄露的问题，这就是所谓的大数据吗？_!_
+6552459636224557326_!_109_!_news_tech_!_大家怎样看“悦花越有”支付？_!_
+6552459412227752196_!_109_!_news_tech_!_目前分析哲学的前沿研究内容是什么？_!_
+6552427257133007364_!_109_!_news_tech_!_马云为什么不收清华大学生，他的回答让每个企业家深思_!_特拉维夫,阿里巴巴集团,北大清华,马云,阿里巴巴
+6552399077529616903_!_109_!_news_tech_!_这才是消费者想要的“共享”_!_保险金,共享资金,经营者,共享经济
+6552328370892308995_!_109_!_news_tech_!_稳定的实力，灵活的规划，在新兴虚拟货币中独迈阿币别具一格_!_迈阿密,迈阿币Mmcoin,虚拟货币,保密措施,区块链
+6552360998559285508_!_109_!_news_tech_!_腾讯的老板马化腾凭什么获得“全球最伟大领袖”的称号？_!_
+6552423942907232515_!_109_!_news_tech_!_HTC的手机质量怎么样，用的人多不？_!_
+6552331341755580680_!_109_!_news_tech_!_为什么中国不能像美国针对华为手机一样也针对美国苹果手机禁售？_!_
+6552373160610627854_!_109_!_news_tech_!_农特产品怎么运用抖音营销？_!_
+6552147528656093443_!_109_!_news_tech_!_携程创始人梁建章呼吁女儿随妈妈姓，淡化男性传宗接代观念。对此你怎么看？_!_
+6552396636864119043_!_109_!_news_tech_!_100年前的世界是如何影响我们今天的生活的？_!_
+6552342919590707716_!_109_!_news_tech_!_马云：阿里并不愿意招清华北大毕业生，他们应该到中小企业去_!_特拉维夫,北大清华,阿里巴巴,马云,名校毕业生
+6552320636885139719_!_109_!_news_tech_!_除了华为、阿里巴巴以外，你还知道哪些称霸世界的中国高新企业？_!_
+6552325556401078536_!_109_!_news_tech_!_当科技遇上历史，将会产生怎样的化学反应？_!_
+6552357945089196552_!_110_!_news_military_!_子弹肌展示_!_
+6552127446471147779_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552451797426897422_!_110_!_news_military_!_天空一声巨响！我国在美眼皮底下击沉万吨军舰，航母后撤1000公里_!_导弹,军舰,安全距离,远程导弹,反舰导弹
+6552490664968323587_!_110_!_news_military_!_中国航母舰载机可以对抗F-22？这个改型有料了_!_F-22,电子战飞机,歼-15,作战条例,电子战机,AIM-120,咆哮者
+6552267799648010765_!_110_!_news_military_!_杜特尔特又说话了，估计会扎特朗普的心！_!_南沙群岛,南海,达沃市,马尼拉,菲律宾总统
+6552387610923237640_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552085006695531011_!_110_!_news_military_!_歼20可以不费吹灰之力！让美国深深的体会到了被打脸的滋味？_!_歼-10,F-35,空空导弹,歼-20战斗机,PL-15,歼-20,苏-35,苏-30,导弹,空军
+6552417494953361677_!_110_!_news_military_!_98K和SKS的子弹初速哪个快？_!_
+6552281359425995011_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552446886303760654_!_110_!_news_military_!_机枪明明打不穿坦克，为什么有的战场上士兵们还坚持射击坦克？_!_
+6550929828973904132_!_110_!_news_military_!_黎以冲突中以色列的哈奈特号护卫舰是怎么被导弹击中的？_!_
+6552488339985924616_!_110_!_news_military_!_不是1只鸟！而是一整群！这才是俄苏30的坠毁原因：运气太差_!_坠毁,苏27,苏35,A-18,战机,苏霍伊,苏30
+6550812929392902413_!_110_!_news_military_!_美国为什么支持叙利亚反对派武装？_!_
+6552227927990206723_!_110_!_news_military_!_子弹出膛瞬间，手腕急速抖动，枪口恰好形成作用于子弹尾部的力量会发生什么变化？_!_
+6552437403787723022_!_110_!_news_military_!_远洋货轮装了十好几米高的集装箱，为什么不会掉到海里？_!_
+6552425325102367245_!_110_!_news_military_!_“寻访退伍老兵，追忆峥嵘岁月”探访慰问活动_!_信阳市,退伍老兵,老兵
+6552275210278535688_!_110_!_news_military_!_国产航母甲板彻底清空，蓄势待发！_!_试航,辽宁舰,航母甲板,首艘国产航母,甲板
+6551767142805537294_!_110_!_news_military_!_知己知彼百战不殆，中美撞机事件后美国开始研发这款无人侦察机_!_无人机,无人侦察机,洛克希德,阿富汗,RQ-170
+6552342557127344647_!_112_!_news_travel_!_“色绿精神”休闲度假的理想之地 有机会一定玩！_!_乌蒙山,金佛山,云贵高原,四川盆地,盘县
+6551651025227350535_!_112_!_news_travel_!_小Bu说：萌爆少女心，淡路岛11米“白色Hello Kitty”巨蛋开幕_!_关西地区,Hello Kitty,淡路岛,海岸边,凯蒂猫
+6552390862125400583_!_112_!_news_travel_!_独家：狂野！狂野！与非洲狮近距离接触大汇总！_!_坦桑尼亚,动物世界,非洲狮,狮子,马赛马拉,雄狮
+6552104407037641230_!_112_!_news_travel_!_重庆8d魔幻之城要你好看，网红不简单，基建狂魔般任性2—铁路篇_!_重庆东站,重庆西站,重庆,云阳县,奉节县
+6552424262433522180_!_112_!_news_travel_!_重庆正在修建一条铁路，预计2020年底建成通车，沿线城市发达了_!_陈独秀,重庆,陈独秀旧居,江津区
+6552486245472141575_!_112_!_news_travel_!_攀爬雕像、随地大小便，“五一”沈阳景区这些现象你怎么看？_!_
+6552328920052531470_!_112_!_news_travel_!_乌衣巷口夕阳斜，下一句是什么？_!_
+6552477935314403843_!_112_!_news_travel_!_去云南的话，忘记大理和丽江吧，这个地方才是云南真正的世外桃源_!_云南,普者黑,世外桃源,丽江,彝族语
+6552449560147395075_!_112_!_news_travel_!_庄浪，你是人间五月的天堂_!_那一年那一天,往日时光,许一世阳光,当代校园文艺,平凉市,青年文星,庄浪,新课程研究,心香无尘,庄浪县,作家协会会员,云崖寺
+6552420250795639048_!_112_!_news_travel_!_立夏到，如何顺应气候养好心？_!_
+6552463610713473288_!_112_!_news_travel_!_人这一辈子会去哪些地方旅行？_!_
+6552268325466931720_!_112_!_news_travel_!_伊亚小镇，堪称圣托里尼亚的精华所在，这里有着全世界最美的日落_!_
+6552364754617238023_!_112_!_news_travel_!_深圳农家乐一日游，亲子活动九龙生态农业园_!_农家乐,生态农业园,一日游,九龙
+6551538023643742724_!_112_!_news_travel_!_比三亚清净，高铁可直达！这个有山有海的小岛，物价便宜又美丽！_!_渔民生活,崂山茶,沙子口,崂山,民宿
+6552418663117357576_!_112_!_news_travel_!_去巴厘岛小心这种“特色服务”，可不要随便“尝鲜”_!_染发剂,海娜纹身,Julie,PPD,巴厘岛
+6552473977158107656_!_112_!_news_travel_!_卢氏县东城学校四年级六班 郭懿<《重渡沟一日游记》>_!_瀑布,重渡沟一日游记,三门峡市,重渡沟,卢氏县
+6552450565786304775_!_112_!_news_travel_!_如何看待，五一假期西安再次火遍抖音？_!_
+6552487487179063811_!_112_!_news_travel_!_中国地理位置优越，世界十大山脉，中国占了五个！我自豪！_!_南美洲,冈底斯山脉,山脉,新南威尔士州,喜马拉雅山脉
+6552254063679898120_!_112_!_news_travel_!_携程承诺不从机票退改签中收取额外费用，标准混乱仍然待解_!_民航局,退票费,机票,退改签,携程
+6551634744252039437_!_113_!_news_world_!_70%的美国人真的连1000块钱都拿不出来，没有任何储蓄吗？_!_
+6536129887235736077_!_113_!_news_world_!_泰国前任总理，英拉年轻时到底有多美，上学时追求者无数！_!_政治学,英拉,他信,泰国,AIS
+6553085902699102733_!_113_!_news_world_!_主播一声“干妈”妇人赏30万再送名表，女儿气的断绝关系_!_江苏省,主播,山东省,打赏费
+6551926485786034701_!_113_!_news_world_!_气愤！美国为阻止中国技术增长，竟然想向100万中国留学生下手_!_中国公民,Qualcomm,特朗普,永久居留,通用电气
+6485242546703302925_!_113_!_news_world_!_雪糕袋子上印有“清真”两个字到底是什么意思？_!_雪糕,清真,伊斯兰教,牛奶,伊斯兰国家
+6550148831638454787_!_113_!_news_world_!_看完印度最有钱的村庄，再看中国最有钱的村庄，对比一个天一个地_!_印度卢比,直升飞机,印度,华西村,Popatrao Pawar
+6551650010985923080_!_113_!_news_world_!_美国独立后，林肯曾想把黑人送回非洲，因一个原因不得不放弃_!_亚伯拉罕·林肯,利比里亚,美国黑人,解放黑人奴隶宣言,林肯,亚伯拉罕
+6536694796205949448_!_113_!_news_world_!_世界各国的“国菜”,法国鹅肝，日本生鱼片，中国国菜我服了_!_好主妇,国菜,生鱼片,鹅肝酱,烤骆驼
+6426884730720878850_!_113_!_news_world_!_在日本，穷人的生活是怎样的？_!_日本经济,日本,精神疾病,穷人,胶囊旅馆
+6551255606404055559_!_113_!_news_world_!_全球最危险6大犯罪集团，排第一位的战斗力凶悍，跟特种部队硬干_!_墨西哥,地狱天使,犯罪集团,山口组,黑帮组织
+6548178525193503236_!_113_!_news_world_!_她15岁做艺妓53岁拍全裸写真76岁跳钢管舞，她是一名总统夫人_!_黛薇夫人,秀雅,钢管舞,印尼,总统府,苏加诺
+6548772809663840782_!_113_!_news_world_!_古代日本派大批女子前往中国，专偷一样东西，之后日本人更加自信_!_日本人,宋朝,正衣冠,中国历史,外交政策
+6499047924725448974_!_113_!_news_world_!_普京前妻结婚了吗？_!_法国人,普京,柳德米拉·普京娜,前妻,俄罗斯
+6550743195448246792_!_113_!_news_world_!_世界上仅有的两个由发达变落后的国家，令人可惜_!_亚洲开发银行,南非,菲律宾,贫民窟,约翰内斯堡
+6484488398592540941_!_113_!_news_world_!_我们的世界有多少个国家？_!_观察员国,联合国会员国,主权国家,巴勒斯坦,俄罗斯
+6553038993334731022_!_113_!_news_world_!_男子在餐车就餐遭拒绝，扬言要曝光乘务员，你怎么看？_!_
+6552374298273645064_!_115_!_news_agriculture_!_集市这些面食摊，摊子不大，但生意非常好！_!_兰州牛肉面,臊子面,小吃摊,热干面,担担面
+6552301475119235597_!_115_!_news_agriculture_!_养鸽子培育鸽子做种和参赛的区别_!_鸽子,留下来做
+6552269044995588616_!_115_!_news_agriculture_!_每亩最高赚七八千 盱眙小龙虾如何“步步高”_!_盱眙,小龙虾,盱眙县,经营者,淮安市
+6551988497169776900_!_115_!_news_agriculture_!_月季花用鱼的内脏水作肥料，花开的旺吗？_!_
+6550772189807772164_!_115_!_news_agriculture_!_五月份果园该做哪些工作？（下）_!_生物杀虫剂,果园,病虫害,花芽分化
+6552221427917914628_!_115_!_news_agriculture_!_产粮大县的田野“四则运算”_!_紫云英,农业,有机肥,宁乡,陈德奇
+6552373680012263950_!_115_!_news_agriculture_!_聊城人快看！我们每天生活的聊城，原来是这个样子！_!_茌平县,贾寨镇,聊城,聊城人
+6552286695662813704_!_115_!_news_agriculture_!_五一猪价继续下跌，小散户不倒闭，猪价就涨不上去？_!_养猪业,养猪场,猪价,小散户,农村
+6552398582987620615_!_115_!_news_agriculture_!_中国哪里的脐橙最好？_!_
+6552298338337686029_!_115_!_news_agriculture_!_就冲这3个条件，红心猕猴桃卖的也要比黄、绿心贵_!_红心猕猴桃,苍溪县,都江堰,核桃苗,苍溪
+6552305352371601924_!_115_!_news_agriculture_!_温室大棚的选择，不求最贵只求更对_!_温室,围护结构,温室大棚,透光率,雪荷载
+6552286059638555139_!_115_!_news_agriculture_!_富县：建成千亩田园综合体 打造黄土高原“稻梦空间”_!_富县,胡家坡村,稻田画,葫芦河,直罗镇
+6552410197199946247_!_115_!_news_agriculture_!_农村低保“政策”大幅度调整！农民：还改？让不让过了！_!_残疾人,重度残疾人,农村
+6552439376326951437_!_115_!_news_agriculture_!_郑州的樱桃熟了吗？听说今年樱桃减产是真的吗，还能采摘吗？_!_樱桃沟,樱桃,小樱桃,大樱桃,紫荆山
+6552299490995339523_!_115_!_news_agriculture_!_现在果园地里墒情不错，可以划沟喂尿素吗？_!_
+6552378113798439431_!_115_!_news_agriculture_!_最优质的大丽花培育基地—甘肃临洮绿洲花卉种植基地_!_大丽花培育基地—甘肃,甘肃,甘肃临洮绿洲花卉种植,临洮,大丽花
+6552302225060790541_!_115_!_news_agriculture_!_老人们常说自己年轻那个时代吃不饱、穿不暖，但那时河里到处都是鱼，地里也有野鸭野兔，为什么不抓来充饥？_!_
+6552038121058337293_!_115_!_news_agriculture_!_吃到绿色安全果蔬不是梦_!_西红柿,西夏镇,沼气池,西华县,转基因
+6552274865649353223_!_115_!_news_agriculture_!_农村这些田里的野草，有的有毒，有的却能吃又健康！_!_贝壳杉烯,农村
+6552486664554414600_!_116_!_news_game_!_斗鱼绝地求生黄金大奖赛F组第一天，numb发挥出色，双连鸡_!_斗鱼,绝地求生,FPS游戏,大奖赛
+6552400078227636740_!_116_!_news_game_!_王者荣耀：小姐姐一分钟收到八个曙光守护者！这样要皮肤真的好吗_!_曙光守护者,小姐姐,王者,女朋友,玩家
+6552302720718471438_!_116_!_news_game_!_王者荣耀里，后羿重做后究竟是增强了，还是不如以前了？_!_
+6552127248114123268_!_116_!_news_game_!_MSI微星游戏本6G独显和7代i7，开挂般的神奇体验！_!_读卡器,笔记本,MSI,GT73EVR 7RD-818CN,微星
+6533164572814082308_!_116_!_news_game_!_韩国有枪战游戏，美国有刀塔游戏，中国有哪些游戏遍布世界？_!_
+6552475218630148365_!_116_!_news_game_!_都说王者峡谷里老公打不过老婆，那为什么王昭君就很难打过李白呢？_!_
+6552492084383711747_!_116_!_news_game_!_王者荣耀：QGhappy在保级边缘挣扎，他们是真的茫然了_!_逆版本,BAN,QGhappy,夏侯惇,QG
+6552257748254327048_!_116_!_news_game_!_怎么看微博大V爆料《王者荣耀》掉帧卡顿等问题，是因为手机品牌保护费交的不够？_!_
+6552319421401006605_!_116_!_news_game_!_《荒野大镖客2》发售前全方位情报汇总，都是各位想知道的！_!_黑水镇,达奇,GTA,Rockstar Games,NBA 2K,帮派,GTA5,马斯顿,荒野大镖客2
+6552466489838928132_!_116_!_news_game_!_东归英雄传用了多少马？_!_
+6552405462489235971_!_116_!_news_game_!_王者荣耀：S11赛季强势英雄榜，张良咸鱼翻身，成中路上分黑马！_!_武则天,王昭君,张良
+6552097744729669891_!_116_!_news_game_!_王者荣耀怎么看双Q大战，以XQ拒绝横扫3:1复仇QGhappy？_!_
+6550132823854940676_!_116_!_news_game_!_小米发烧，新品生为竞技，6G+64G+4000mAh+液冷，黄牛加价1000_!_游戏手柄,小米,游戏,小米公司,小米手机
+6552373643907695107_!_116_!_news_game_!_孩子沉迷《堡垒之夜》！卫报：游戏没错，错的是时间_!_Hancock,怪奇物语,堡垒之夜,电子游戏,文化部长,网络设计,色情游戏
+6552465426276680196_!_116_!_news_game_!_DNF：强化上12连续失败怎么办？这位大佬的做法太霸气！_!_网络游戏,游戏,玩家,武器,DNF
+6552275289756402183_!_116_!_news_game_!_游戏时光本周热门：去完北欧，今年还有西部_!_荒野大镖客救赎2,格斗游戏,掌机,豪勇七蛟龙,玩家,武侠,游戏机实用技术,战神
+6552284874466656519_!_116_!_news_game_!_《王者荣耀》梦奇和铠单挑谁更厉害？_!_
+6552360181991211277_!_116_!_news_game_!_《王者荣耀》的哪个英雄或者哪个技能最让你怀念呢？_!_
+6552296988992340487_!_116_!_news_game_!_英雄联盟：下个版本QDC玩法又回来了，拳头屡次削弱都没用_!_英雄联盟,无尽之刃,玩法,游戏,EZ
+6552353371548811780_!_116_!_news_game_!_有了它之后，吉利服变成了自我欺骗，韦神女友都看不过去了！_!_吉利服,平底锅,Long,韦神,游戏
+6552412827095662861_!_101_!_news_culture_!_家居装饰画选什么题材，鱼和虾哪个更合适？_!_
+6552466761776627971_!_101_!_news_culture_!_中国古代的刑法有哪些？_!_
+6552472109417759239_!_101_!_news_culture_!_初夏，衡阳县船山广场倒影如幻，美如油画_!_
+6552360580542365955_!_101_!_news_culture_!_上联：遍地鲜花摇落地，如何对下联？_!_
+6552440883734643204_!_101_!_news_culture_!_河阳巨里，楼里左镇——探访邢台千年古村后楼下_!_隆兴寺,邢台七里河,楼下村,古建筑,七里河
+6552329673576022535_!_101_!_news_culture_!_七绝 平水韵《潮州单丛茶》_!_潮州单丛茶,平水韵,入声字
+6552191574636232967_!_101_!_news_culture_!_假如秦始皇没有统一中国，后来中国会统一吗？_!_
+6552284549563285773_!_101_!_news_culture_!_刘备有四个儿子，为何却让资质平平的刘禅继位？_!_
+6552481484584780296_!_101_!_news_culture_!_国画艺术走近生活 名家陶瓷绘画交流会5月6日举办_!_书画院,王福顺,山水画,书画家,国画艺术
+6552218089373565197_!_101_!_news_culture_!_上联：山水不曾随我老，如何对下联？_!_
+6552319432952119556_!_101_!_news_culture_!_玉笛吹来美人风，如何赐句？_!_
+6552308280016765187_!_101_!_news_culture_!_上联: ，儒非儒。道非道。自称老子。有什么下联可对？_!_
+6549311977276047630_!_101_!_news_culture_!_金庸武侠世界中的门派为何大部分超不过三代就会衰落？_!_
+6552316668792537347_!_101_!_news_culture_!_上联：江中鱼东游西藏，下联怎么对？_!_
+6552253667452387843_!_101_!_news_culture_!_现实派艺术家——Jeremy Lipking油画作品_!_油画,加利福尼亚州,艺术家
+6552392852763050504_!_101_!_news_culture_!_项王故里写丹青_!_项王故里,宿迁市
+6552424912387047687_!_101_!_news_culture_!_下联:周六日双休。上联怎么对？_!_
+6552425657677119747_!_101_!_news_culture_!_上联：江中鱼东游西藏，下联怎么对？_!_
+6551635605707555335_!_101_!_news_culture_!_抹额，你知道它还有御寒的功效吗？发展到现代成为一件美丽的配饰_!_王熙凤,抹额,续汉书·舆服志,席上腐谈,红楼梦,新唐书·娄师德传
+6552433201715872264_!_102_!_news_entertainment_!_谢娜产后复出魅力不减，粉丝数量过亿，获两项吉尼斯纪录_!_娜娜,鱼尾纹,吉尼斯,快乐大本营,谢娜
+6552376067171025411_!_102_!_news_entertainment_!_《我是歌手》中选手出场费曝光，张杰比韩红还要高？他竟要220万_!_林忆莲,韩红,蒙面唱将,我想和你唱2,歌手,凉凉,中国好声音,王牌对王牌,张碧晨,跑男,出场费,湖南台
+6552380069715640836_!_102_!_news_entertainment_!_《快乐大本营》谢娜C位复出，遇到3位长腿女星，彻底露怯了_!_快乐大本营,谢娜,何炅,蔡少芬,超时空同居,佟丽娅,奚梦瑶
+6552319314727272973_!_102_!_news_entertainment_!_谢娜回归《快乐大本营》过37岁生日，快乐家族集体送祝福_!_快乐家族,吴昕,快乐大本营,杜海涛,谢娜,何炅
+6552433784677990920_!_102_!_news_entertainment_!_第29届格拉德媒体颁奖典礼在纽约举行_!_格拉德,29届,东方IC,东方ic,届格拉德媒体颁奖典礼
+6552424870267847181_!_102_!_news_entertainment_!_身为K-pop领袖，东方神起希望能给后辈带来正面影响_!_K-pop,韩国,赵容弼,演唱会,东方神起
+6552032879403074055_!_102_!_news_entertainment_!_“票房灵药”周冬雨_!_金城武,喜欢你,同桌的你,春风十里不如你,周冬雨姐姐,麻雀,爱情片,山楂树之恋,冰河追凶,七月与安生,不完美女孩,安妮宝贝,我们相爱吧,林更新
+6552388077338231048_!_102_!_news_entertainment_!_为什么刀郎越隐越红？_!_
+6552308278523593219_!_102_!_news_entertainment_!_很多香港明星讲普通话艰难，台湾明星却无此窘境，原来是因为这些_!_台湾,普通话,香港,东方IC,国语
+6552415143521681927_!_102_!_news_entertainment_!_谢娜人缘有多好？看到何炅这一句话就知道啦_!_田亮,叶一茜,谢娜,何炅,章子怡
+6552271738112049672_!_102_!_news_entertainment_!_难得夸人的梁朝伟，本想夸赞一下她的演技，结果却夸得她哭笑不得_!_孙俪,大众电影百花奖,郑晓龙,Angelababy,摆渡人,梁朝伟,甄嬛传,芈月传,寻龙诀
+6552269231142994190_!_102_!_news_entertainment_!_何超雄是身价数十亿的千金，为什么没人敢娶？_!_
+6552402653958111752_!_102_!_news_entertainment_!_黑寡妇第一毫无悬念！漫威8位最火的超级女英雄排行_!_复仇者联盟2 ：奥创纪元,漫威,荷里活,钢铁侠2,混血儿,凯伦·吉兰,复仇者3,螳螂
+6552265030937608707_!_102_!_news_entertainment_!_超能造梦第二季什么时候出？超能造梦第二季在哪里可以看_!_高林豹,夏至未至,鲜肉老师,花千骨,丁一,郑合惠子,刘芮麟,无心法师,超能造梦
+6552239835254358286_!_102_!_news_entertainment_!_《甄嬛传》中皇后为何没在宫外除掉甄嬛？_!_
+6552296463102116355_!_102_!_news_entertainment_!_王俊凯日本奈良被偶遇，戴着眼镜喂小鹿，网友：看成王俊凯喂鹿晗_!_漫画,日本,王俊凯日本,美少年,小鹿
+6552403702743826958_!_102_!_news_entertainment_!_中国内地NINE PERCENT男子组合全球首场粉丝见面会在上海隆重开锣_!_蔡徐坤,Justin,陈立农,东方IC,偶像练习生,范丞丞
+6552424250077085966_!_102_!_news_entertainment_!_你喜欢看广西卫视的扶贫节目《第一书记》吗？_!_
+6552430335211077892_!_103_!_news_sports_!_权健是做什么的，为什么那么有钱？_!_
+6552285777055711758_!_103_!_news_sports_!_悲剧又要重演，猛龙连续三场失误上双，詹姆斯超帕克成现役第一_!_猛龙,德罗赞,上半场,骑士,范弗里特
+6552480652380340494_!_103_!_news_sports_!_詹姆斯现在的绝杀球能力在联盟中是什么水准？_!_
+6552353490075648515_!_103_!_news_sports_!_下一个世界前十竟然是他？打败雅桑克莱和考的人说先过我这关_!_搏击,雅桑克莱,法比奥,昆仑决,铁英华
+6552375226024329476_!_103_!_news_sports_!_中超第9轮，国安对上港，比赛最后阶段裁判的判罚引发争议，你认为判罚是否存在问题？_!_
+6552340270359249160_!_103_!_news_sports_!_为什么欧冠决赛没有国内媒体去现场解说，成本很高吗？_!_
+6552465360086368771_!_103_!_news_sports_!_中超联赛第9轮，重庆斯威0-1天津权健_!_中超联赛,天津权健,东方IC,重庆斯威0-1,9轮
+6552307184401645832_!_103_!_news_sports_!_刘伟解说乒乓球比赛的水平如何？_!_
+6552486382638465540_!_103_!_news_sports_!_亨德森：荣幸为克洛普效力，皇马是强大的队伍_!_Goal.com,Desperado,欧冠决赛,皇马,亨德森,克洛普
+6552319762494390542_!_103_!_news_sports_!_如果骑士火箭进入总决赛，谁的胜算大？_!_
+6552316456850161934_!_103_!_news_sports_!_从个人竞技状态来看，三个阶段的詹姆斯，哪个最强？_!_
+6552449635942662660_!_103_!_news_sports_!_骑士总冠军！地球人谁能阻挡詹姆斯？史上最佳就是他！打服所有人_!_猛龙,德罗赞,阿奴,詹姆斯,骑士
+6552302851983409677_!_103_!_news_sports_!_詹姆斯绝杀，骑士3比0，猛龙怀疑人生_!_上半场,多伦多,NBA,詹姆斯,猛龙
+6552470960530784526_!_103_!_news_sports_!_骑士和步行者战成抢七险胜，而猛龙即将被横扫，步行者跟猛龙的区别在哪里？_!_
+6552333784631476493_!_103_!_news_sports_!_骑士G3詹姆斯8秒绝杀猛龙，为什么网友都心疼起了德罗赞和洛瑞？_!_
+6552344263059833352_!_103_!_news_sports_!_回顾科比经典：幼科单挑基德2.3秒绝杀！科比霸气动作庆祝_!_替补席,太阳队,季后赛,基德,科比
+6552477436729098510_!_103_!_news_sports_!_西班牙国家德比大名单最新出炉，无欲无求之下，皇马巴萨决定死磕，对此你怎么看？_!_
+6552420398196064526_!_104_!_news_finance_!_一家信用卡逾期五个月，他行信用卡会封卡降额吗？_!_
+6552325196118753795_!_104_!_news_finance_!_创业你没钱？不创业你更没钱！2018那些你不知道的商机_!_创业者,诸葛亮,合伙创业,马云,大健康,阿里巴巴
+6552396662168355080_!_104_!_news_finance_!_A股为何不实行T+0交易制度？_!_
+6552127239201227268_!_104_!_news_finance_!_唯独一家不用明星代言的手机，就连股神巴菲特都力挺！_!_巴菲特,苹果股票,伯克希尔,iPhone,股苹果股票
+6552351570275598862_!_104_!_news_finance_!_炒股天才的肺腑独白，股票书上无法学到的精髓，练到极致就是绝活_!_主力,MACD,股价,骗线,开盘价,技术指标
+6552339558711689736_!_104_!_news_finance_!_那些可以一把梭的股票 都有啥特征_!_股票,保千里,否极泰来,立于不败之地,金刚玻璃,高斯贝尔
+6552408759744856328_!_104_!_news_finance_!_庄家出货不是在高位出货的吗，为什么要砸盘到跌停出货呢？_!_
+6552377528424595716_!_104_!_news_finance_!_飞机票退改签手续费谁说了算？是否属于霸王条款？_!_
+6552356933322408205_!_104_!_news_finance_!_美国国债也是借新钱还旧钱，是不是庞氏骗局？_!_
+6548659662533689603_!_104_!_news_finance_!_为什么有钱人理财能10倍100倍增长，而我们理财原地踏步10年不动呢？_!_
+6552355950643118343_!_104_!_news_finance_!_巴菲特说股票交易是一个简单理念的坚持，你是怎么看的？_!_
+6552484429309149709_!_104_!_news_finance_!_2018.5.7经济解读_!_中国银行,上市规则,华泰证券,三一重工,药明康德,兴业银行,QDII,瑞银,MSCI,京东方,兴源环境,蔡奇,亚夏汽车,卡拉奇,国泰君安,大唐电信,QFII,方正证券,RQFII,佳士科技
+6552460814324859144_!_104_!_news_finance_!_当你所做的一切都因为“钱”而止住时，身无分文，该如何解决眼前的困境？_!_
+6552292517239324936_!_104_!_news_finance_!_为什么很多人炒股同时用很多台电脑？_!_
+6552407342028161283_!_104_!_news_finance_!_有哪些不被看好的行业正在崛起？_!_
+6552386415248802318_!_104_!_news_finance_!_有人说买房贷款越多越好、还款时间越久越好，真的是这样吗？_!_等额本息,银行,买房,购房者,等额本金
+6552395941767283207_!_104_!_news_finance_!_庄家在买二买三挂大买单托单是想干什么？_!_庄家,股票,散户
+6552158568991162887_!_104_!_news_finance_!_一批重磅新规今天正式实施_!_沪港通,快递暂行条例,抗癌药,海南,第四套人民币
+6552369878420947464_!_104_!_news_finance_!_广州红酒进口报关报检_!_产地证,红酒生产工艺流程,成份鉴定分析表,红酒,进口物流,进口成本,进口代理,红酒进口报关
+6552371077543100942_!_106_!_news_house_!_又涨了！淄博市最新5月二手房价曝光，快看看你家又涨了多少？_!_㎡盛世,安居客,二手房,房价,二手房均价
+6552290538131816711_!_106_!_news_house_!_大家身边的朋友都有几套房，你怎么看房地产的未来十年？_!_
+6552381931806261518_!_106_!_news_house_!_马云又出惊人言论，8年后房子最不值钱，他的话可信吗？_!_
+6552290185021751559_!_106_!_news_house_!_职工在单位工作期间交纳公积金，下岗后如何在社会上缴纳住房公积金？_!_
+6552496280218632456_!_106_!_news_house_!_2018年随着二胎政策放开，三四线城市房价还在上涨，这两者有何关系，房价上涨的原因有哪些？_!_
+6552342188569657860_!_106_!_news_house_!_在中国一辈子不买房，一直租房现实吗？_!_八通线,二房东,外地人,房贷,香河,斯里兰卡,买不起,房价
+6552464491311792392_!_106_!_news_house_!_重庆主城区的房子还有增值空间吗？_!_
+6552406627985654276_!_106_!_news_house_!_上海4月份住宅供应涨84.5% 成交量创8年来新低_!_商品住宅,中原地产,上海
+6552317944670454280_!_106_!_news_house_!_政府卖地收入暴涨83.4%！这几个城市正在悄然崛起_!_土地出让金,房价,城市土地,楼市,消化周期
+6552342240977486349_!_106_!_news_house_!_合肥商圈人气惨淡，又有2家商场即将倒闭，商场内一片萧条_!_太阳城,商业街,女人街,温莎国际广场,合肥
+6552411186158109197_!_106_!_news_house_!_深圳房价每平米下跌16元幕后真相？_!_房价,二手房,深圳,深圳新房,深圳房价
+6552367250232312324_!_106_!_news_house_!_一句话楼市：天河积分入学公办学位逾800个3500个来穗家庭申请_!_三横两纵,穗人员,广汕路,积分入学申请,智慧城
+6552382801931403524_!_106_!_news_house_!_在五六线小县城现在要不要买房？_!_
+6552474130128568580_!_106_!_news_house_!_现在能筹到30-40万，惠州买房值得吗？_!_
+6552252819343147278_!_106_!_news_house_!_为什么房价低的时候没人买，房价猛涨大家却都抢着买？_!_
+6552347962209468679_!_106_!_news_house_!_房贷利率上涨，为何银行对房贷出现180度态度转变？_!_
+6552405891910467843_!_106_!_news_house_!_副省级城市与计划单列市有什么区别？_!_
+6552310361599508749_!_106_!_news_house_!_成都物业保洁的详细内容是怎样的呢？_!_
+6552477965559529732_!_106_!_news_house_!_运动会都有什么项目？_!_
+6552374636493931022_!_107_!_news_car_!_汽车大灯发黄，老化，划痕还多？别担心，这个小妙招教你手动翻新_!_汽车音响,多功能,音频线,防滑垫,汽车大灯
+6552346171082277384_!_107_!_news_car_!_吉利拼了，新车逼格完胜牧马人，配3.5升加四驱，仅售15万_!_吉利帝豪,SUV,牧马人,帝豪EX9,沃尔沃,EX9
+6552324097995440653_!_107_!_news_car_!_驾考科目三靠边停车，4步骤教你精准找到靠边停车30公分_!_右边缘线,踩刹车,靠边停车
+6552341004073042190_!_107_!_news_car_!_日产骐达智行版12.7万落地贵吗？_!_
+6552437741592773127_!_107_!_news_car_!_幸亏我知道得早，做到这几点，发动机用20年都不是问题！_!_曲轴箱,热车,机械能,积碳,发动机
+6552373525733179661_!_107_!_news_car_!_汽车手刹不太灵跟刹车片有没有关系？_!_
+6552416091828650504_!_107_!_news_car_!_长安CS75这款车怎么样，值得买吗？还是听听车主是怎么说的吧！_!_rx5,长安汽车,比亚迪宋,s330,长安,变速箱,胎压监测,比亚迪
+6552411106822849032_!_107_!_news_car_!_长城发飙了！推出新车霸气不输汉兰达，售价15万起，途观慌了_!_哈弗,SUV,汉兰达,哈弗H6,长城,液晶屏
+6552466560139657479_!_107_!_news_car_!_途观L和x1，哪个好？_!_
+6552430135532847374_!_107_!_news_car_!_如果给你三十万，你会选择什么车？_!_
+6552321010786370056_!_107_!_news_car_!_海马的设计师终于开窍了！这才是颜值担当！_!_海马S5,海马,海马汽车,运动感,马自达,保险杠
+6552331335455736333_!_107_!_news_car_!_雨天刮着雨刮开车，车窗越刮越脏还有异响！一个新方法，干净隔音_!_洗车机,密封条,汽车轮胎,引擎盖,无骨雨刷
+6552414196506558734_!_107_!_news_car_!_50岁的男人适合开哪款车？_!_
+6552475360460538372_!_107_!_news_car_!_聊聊我所知道的张雪 和他的凯越500X_!_摩托车,凯越,湖南电视台,张雪,牛摩网
+6552476031331074564_!_107_!_news_car_!_二手车评估师教你如何购买二手车？_!_二手车,58同城,二手车评估师,日产轩逸
+6552466715387625741_!_107_!_news_car_!_想换个30万左右的车，现在开科鲁兹，有什么好推荐的？_!_
+6552302504116224520_!_107_!_news_car_!_哈弗H6售价9.3万元起 欢迎到店品鉴_!_哈弗H6,万2017,换代红标,自动两驱,哈弗,万0.30
+6552453483239309831_!_107_!_news_car_!_路上爆胎怎么办？备上这款中国造“换胎神器”，使用省力更方便_!_体验汽车倒车盲点镜,MP3,柔软汽车坐垫新款,汽车轮胎,遮阳板
+6552249931627757832_!_107_!_news_car_!_本田xrv现在优惠多少？_!_
+6552311345298014472_!_108_!_news_edu_!_在南京艺术学院读书，前景如何？_!_
+6552407720639595016_!_108_!_news_edu_!_小记——一封迟来的信_!_朱自清
+6552275960664687111_!_108_!_news_edu_!_毕业季来临 台生“登鹭”实习忙_!_厦门,台湾,台青创客家,台生,台生实习
+6552301106658017796_!_108_!_news_edu_!_看北师大学子用书法展示总书记的青年说_!_各界优秀青年代表座谈,书法,中国政法大学,核心价值观,苟日新
+6552355334638273032_!_108_!_news_edu_!_北京西城区小升初报名通知：外区外省学生回西城就读须有房有户口_!_申请表,公有住宅租赁合同,户口簿,小升初,房屋所有权证书,西城区,不动产权证书,法定监护人,出生医学证明
+6552488179985809667_!_108_!_news_edu_!_我很非常努力，平时的考试成绩也可以，可为什么一到竞赛就考不好？_!_
+6552429464477761799_!_108_!_news_edu_!_课后服务由盯班老师放学而不是由班主任放学，安全隐患有多大？_!_
+6552440662891954701_!_108_!_news_edu_!_双语阅读：与人交谈时 为什么很难保持眼神接触？_!_eye contact,Japan,University,日本京都大学,眼神接触
+6552361602232877571_!_108_!_news_edu_!_北大校长因演讲读错“鸿鹄”发音致歉 但好像没人买账……_!_北大校长,林建华,道歉信,实践论,北京大学,对立统一,矛盾论
+6552291329039139085_!_108_!_news_edu_!_西安小升初电脑摇号加面试，要不要报面试培训班？_!_
+6552485367344267534_!_108_!_news_edu_!_基础差的同学如何复习考研数学？_!_
+6552301039393964292_!_108_!_news_edu_!_郑州大学与中国科学技术大学，哪个更强？_!_
+6551993176385651214_!_108_!_news_edu_!_为什么要上大学？这是我见过的最好的回答_!_奢侈品,音乐会,企业家,人文素养
+6552354142944231944_!_108_!_news_edu_!_亲子教学—为了天使的明天_!_康复,亲子教学,亲子班,行为习惯,助听器
+6552455642748027396_!_108_!_news_edu_!_重磅！许昌这3名学生入选教育部名单，有望拿下国家级奖学金！_!_教育部,宋庆龄奖学金,苌庄乡,获奖候选人,中国福利会
+6552311997768139021_!_108_!_news_edu_!_求证：为什么2017=2018？_!_
+6552321548982682126_!_108_!_news_edu_!_教育者论坛探讨创新人才核心素养_!_上海市,周洪波,教育者论坛,人文素养,上海培佳双语学校
+6552299396027908365_!_108_!_news_edu_!_孩子学习成绩全班倒数，为什么老师不能抽时间为孩子补补课，老师不应该帮孩子提高成绩吗？_!_
+6552347393646395662_!_108_!_news_edu_!_有什么最简炼的语言或词汇能摡括人生的整个里程？_!_
+6552020162860548611_!_109_!_news_tech_!_世界首富告诉我们，这么做就对了——千万别留遗憾！_!_王健林,贝佐斯,马化腾,比尔·盖茨,长江实业
+6552493007520661764_!_109_!_news_tech_!_怎样使用微信收付款？_!_
+6552452608223609347_!_109_!_news_tech_!_梦想是弱者的精神支柱，腾讯要这玩意儿干啥_!_从头再来,天方夜谭,腾讯没有梦想,马化腾,短视频,腾讯
+6552270488880546317_!_109_!_news_tech_!_小米百货模式已形成，雷军要当杂货铺老板吗？_!_智能手机,小米,雷军,互联网服务,互联网公司
+6552458301152428551_!_109_!_news_tech_!_独角兽的标准_!_CDR,白云山,存托凭证,阿里,互联网
+6552269691824374023_!_109_!_news_tech_!_腾讯市值超过阿里，但为什么在各种场合上，总是感觉阿里格局更高？_!_
+6552306305267138829_!_109_!_news_tech_!_建一个微信小程序商城需要多少钱？_!_
+6552476196146249987_!_109_!_news_tech_!_微信删除并拉黑了对方，我现在再把他添加回通讯录，对方会不会有提示？_!_
+6552263145954476295_!_109_!_news_tech_!_现在iPhone X销量如何？_!_
+6552139848113717518_!_109_!_news_tech_!_我这个老电脑怎么才能玩《吃鸡》？_!_
+6552301564621488388_!_109_!_news_tech_!_当前的大环境下，想学一门编程语言，该学什么编程语言呢？_!_
+6552400897446511108_!_109_!_news_tech_!_一个腾讯QQ号到底能加多少好友呢？_!_腾讯电脑管家,添加好友,qq好友,腾讯,QQ
+6552478260305854734_!_109_!_news_tech_!_这个电脑配置值多少钱？_!_
+6552439594976018692_!_109_!_news_tech_!_你是支持网约车还是支持出租车？_!_
+6552290872526897411_!_109_!_news_tech_!_pos机流量卡怎样充值续费？_!_
+6552285041836163332_!_110_!_news_military_!_军演时来敌人了怎么办？_!_
+6552408544426066189_!_110_!_news_military_!_为什么各国坦克到中东军队都会变烂？_!_
+6552480330475897102_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552329871895298308_!_110_!_news_military_!_美国声称“没有输过任何一场战役”，但是为何美国却输掉了越南战争？_!_
+6552268608137855240_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552417222919192836_!_110_!_news_military_!_印度为什么不装备战略轰炸机？_!_
+6552441062583959812_!_110_!_news_military_!_为什么子弹不设计成击中爆炸这种效果，威力不是更大吗？_!_
+6552488746204266760_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552382308668670215_!_110_!_news_military_!_如果发生世界大战，你会怎么办？_!_
+6552450184696037640_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552292285696966915_!_110_!_news_military_!_叙利亚的前途在哪里？_!_
+6552325884303376643_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552454969482543619_!_110_!_news_military_!_再遭重大丑闻袭扰！英媒感叹：皇家海军辉煌什么时间回来？_!_英国皇家海军,航母,日不落帝国,皇家海军,航空母舰
+6552381344171688199_!_110_!_news_military_!_如何评价墨索里尼？_!_
+6552280714761470222_!_110_!_news_military_!_为什么导弹能打飞得快的飞机，却难打跑得慢的航母？_!_
+6552298587210907912_!_110_!_news_military_!_俄罗斯作为世界强国，在近代历史上为何没和英法一样占据海外殖民地？_!_
+6551566505052471555_!_110_!_news_military_!_抗战剧里士兵端着捷克轻机枪一边前进一遍战斗，现实中有没有这个可能？_!_
+6552132537605423629_!_112_!_news_travel_!_远离喧嚣却投入到另一种喧嚣中，游古北水镇见闻_!_云冈石窟,月老祠,大都市,悬空寺,古北水镇
+6552323172736172548_!_112_!_news_travel_!_携程7300元的机票退票只得500元，还有6800元哪去了？_!_深圳航空,退票费,机票,退票,携程
+6552319443865698819_!_112_!_news_travel_!_318国道沿途——四川安居区_!_卧佛院,安居区,遂宁市,白马镇,安岳县
+6552438764596101384_!_112_!_news_travel_!_旅行教会了你什么？_!_
+6552310493908828675_!_112_!_news_travel_!_呼伦贝尔心之所往，来一场说走就走的旅行_!_呼伦贝尔
+6552495766215066120_!_112_!_news_travel_!_沂水风光邀您共赏-为家乡_!_江北乡,烟雨梦,蒙山沂水,风光
+6552459941603443207_!_112_!_news_travel_!_走！露营去！长沙此地人少景美_!_长沙,玉皇殿,石柱峰
+6552369574799474948_!_112_!_news_travel_!_武汉有哪些24小时不打烊的店？_!_
+6552368693727199747_!_112_!_news_travel_!_最古老的千年栈道，盘踞在悬崖峭壁上，堪称鬼斧神工！_!_宁武,栈道,悬空寺,古栈道,大佛殿
+6552399903807504904_!_112_!_news_travel_!_旅游文化：2018中国旅游景点大全！2018年旅游风景区有哪些？_!_王家大院,皇城相府,山西省,客家人,康熙字典,福建土楼
+6552428847717941773_!_112_!_news_travel_!_仁寿光明路红绿灯、高家路烂、噪音扰民……这些问题已解决_!_混疑土,三岔路口,光明路,噪音扰民,红绿灯
+6552326866479350019_!_112_!_news_travel_!_如何规划贵州旅游？_!_
+6552433023705416200_!_112_!_news_travel_!_旅游旺季，来黄鹤楼旅游的外地游客不少，看了都说好_!_
+6552334283757847048_!_112_!_news_travel_!_韩寒带火了东极岛，却不知舟山还有座文艺范十足的小岛_!_冷山,海鲜,梭子蟹,龙头鱼,圣托里尼,带鱼,白沙岛,海钓
+6552440352245022980_!_112_!_news_travel_!_九寨沟开放后有去过的吗怎么样？_!_
+6552432459777049096_!_112_!_news_travel_!_中国最尴尬的地级市，下辖县比自己有名，连机场高铁都在县里_!_泰宁,天鹅洞群,沙县,沙县小吃,三明
+6552380528195010823_!_112_!_news_travel_!_西安好玩还是成都好玩？_!_
+6552294163881132548_!_113_!_news_world_!_加沙发生爆炸致死6人_!_加沙地带,阿什拉夫,卡德拉,卡桑旅,哈马斯
+6552483839044747534_!_113_!_news_world_!_特朗普为什么可以随意废除此前美国政府达成的各种协议？_!_
+6552455511051075843_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552460576457490957_!_113_!_news_world_!_叙利亚战争：独裁与民主_!_叙利亚战争,逊尼派,阿萨德,巴沙尔·阿萨德,什叶派
+6552379507863453966_!_113_!_news_world_!_医改中的虹吸现象是指什么？_!_
+6552253201830117896_!_113_!_news_world_!_哈里王子婚礼更多细节曝光 将首次见到未婚妻父亲_!_婚礼,哈里王子,温莎城堡,金装律师,马克尔,伊丽莎白二世
+6552447842844148227_!_113_!_news_world_!_柬埔寨集体中毒事件已致10死 疑因杀虫剂污染水源_!_杀虫剂,桔井省,柬埔寨,中毒事件
+6552442448725934340_!_113_!_news_world_!_为什么韩国总统没有好下场？_!_
+6552434758717014275_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552488707994157571_!_113_!_news_world_!_菲媒：杜特尔特将在菲律宾保持“严厉”统治_!_禁毒战,亚洲开发银行,菲律宾,杜特尔特,棉兰老岛
+6552315879923646733_!_113_!_news_world_!_如果特朗普遭弹劾下台，谁会来接替他当总统？_!_
+6552242695098597639_!_113_!_news_world_!_作为俄罗斯总统，普京上台以来有哪些失误？_!_
+6552358491808334088_!_113_!_news_world_!_为什么日本人很少有艾滋感染者？_!_
+6552289982692721160_!_113_!_news_world_!_男子体重到达500斤，却娶了一个90斤的媳妇，结果表示很郁闷_!_体重,男子
+6552307322083869191_!_113_!_news_world_!_世界上最危险的20座机场，危险度堪称“死亡机场”！_!_机场,国际机场,飞机跑道,南极洲,巴拉迪
+6550822433387970830_!_113_!_news_world_!_美国为什么要出口垃圾？_!_
+6552284969853518083_!_113_!_news_world_!_如果美国跟以色列也直接介入叙利亚战争，俄罗斯跟伊朗会怎样？_!_
+6552463720436465927_!_113_!_news_world_!_星空琴行创始人周楷程依旧失联，自述为何选择“消失”？_!_
+6552314447912763661_!_113_!_news_world_!_美国下一届大选特朗普会下课吗？_!_
+6552483840105923075_!_113_!_news_world_!_来自领国的软妹子_!_北京奥运会,叶夫根尼娅·卡纳耶娃,软妹子,艺术体操,俄罗斯
+6552438360693015048_!_115_!_news_agriculture_!_12X13米三层带车库别墅，实用性极强外形简约不简单，村长都点赞_!_别墅,农村
+6552410880481427975_!_115_!_news_agriculture_!_农村种植春玉米防治蓟马虫害，其影响产量不能忽视_!_农村,春玉米,玉米螟,玉米
+6552355030366683656_!_115_!_news_agriculture_!_这个野菜在农村都浪费掉了，城里人却当成宝！_!_蛤蟆,荔枝草,中药材,蛤蟆草,癞蛤蟆
+6552025049413452291_!_115_!_news_agriculture_!_雨水频繁迎立夏，及时开展病虫害的防治工作！_!_烯啶虫胺,吡虫啉,病虫害,蚜虫,噻虫嗪
+6551928131786113283_!_115_!_news_agriculture_!_玉米5月首周拍卖结束，怎么看待后市行情？_!_
+6552449087709381127_!_115_!_news_agriculture_!_吃得更放心！农村“坝坝宴”管理也有新招啦~_!_坝坝宴,APP,流动厨师,农村群宴,食安办
+6552381803976458760_!_115_!_news_agriculture_!_农村人招待客人，虽然清淡简单，可满满都是心意！_!_四季豆,主人家,麻辣烫,客人,农村
+6552334086478758413_!_115_!_news_agriculture_!_农村常种植四种的经济作物_!_红薯叶,农村,经济作物
+6552295463511392782_!_115_!_news_agriculture_!_阜南县赵集镇中心学校、刘集小学和大聂九年制教师顶风冒雨-扶贫_!_阜南县赵集镇,教师
+6552265534207951112_!_115_!_news_agriculture_!_喷施百垄通后多久可以栽地瓜？_!_
+6552398865201365518_!_115_!_news_agriculture_!_稻田套养小龙虾做到这六点，不赚钱都不行_!_肥水,小龙虾,池塘,养殖,饲料
+6552445247744377348_!_115_!_news_agriculture_!_这恶魔据闻来自美国，本是好意引进，却泛滥成灾，农民苦不堪言！_!_东方网,美国
+6552025996709593603_!_115_!_news_agriculture_!_辣椒幼苗茎中部萎蔫缢缩，咋回事？怎么治_!_甲霜灵,幼苗,苗床,福美双,猝倒病
+6552263976187593219_!_115_!_news_agriculture_!_玉米行情：2018年5月6日玉米价格行情预测_!_吨山东省,昌乐县,襄汾县,长岭县,玉米价格,莒南县
+6552130959213330957_!_115_!_news_agriculture_!_后申遗时代的哈尼梯田 “稻鱼鸭种养”欲破弃耕困窘_!_红米,元阳,文化遗产,红河哈尼梯田,哈尼族,老虎嘴,元阳县,鸭子
+6552436477958029838_!_115_!_news_agriculture_!_如果家有院子千万别浪费！今年都兴种这种果树，好种易活年年结果_!_蓝莓,山竹,有机质,樱桃树,桃树苗
+6552265535730483716_!_115_!_news_agriculture_!_土地确权中哪些家庭成员属于共有人需要登记？_!_出嫁女,土地承包,家庭成员,新生儿,土地承包法,承包地
+6552437384955298308_!_115_!_news_agriculture_!_湖南省新化县洋溪镇欧式二层半小别墅特别大气_!_大气湖南省新化县洋溪镇欧式二层,湖南省,新化县,别墅特别,特别大气湖南省新化县洋溪镇欧式
+6552343173396431374_!_116_!_news_game_!_全场空勾也让人惧怕三分！王者荣耀遇到钟馗简直是斗智斗勇！_!_钟无艳,钟馗,前中期,武则天
+6552375128515150083_!_116_!_news_game_!_怎么看微博大V爆料《王者荣耀》掉帧卡顿等问题，是因为手机品牌保护费交的不够？_!_
+6552294359587357197_!_116_!_news_game_!_打麻将你不懂技巧？麻将大师教你5个技巧口诀 打牌不会再输_!_麻将,棋牌游戏,打麻将,麻将游戏
+6552436537714278663_!_116_!_news_game_!_steam是否毁灭了中国游戏市场？_!_
+6536181431696097539_!_116_!_news_game_!_英雄联盟喜欢玩辅助的是什么样的人？_!_
+6552470083795419405_!_116_!_news_game_!_你充过钱的游戏中，哪些游戏是你最后悔的？_!_
+6552390157528465933_!_116_!_news_game_!_6日热点新闻：可口可乐日本又出新品，可以冻起来吸着喝_!_s11,可口可乐日本,巅峰赛
+6552424912739369229_!_116_!_news_game_!_玩《英雄联盟》时，总误触ctrl键，怎么解决？_!_
+6552252343495164423_!_116_!_news_game_!_DNF：体验服光环装备改动，部分副本中无效，已经彻底凉了！_!_85级,卡特勒,职权臂章,光环装备,卡勒特
+6552265177641779470_!_116_!_news_game_!_怎么看微博大V爆料《王者荣耀》掉帧卡顿等问题，是因为手机品牌保护费交的不够？_!_
+6552277852958491144_!_116_!_news_game_!_5日游戏热点回顾：王者荣耀：玩家求官方加强猴子，别改李白，策划一一答复_!_猴子,孙亚龙,李白,LOL,Uzi
+6492997018913341709_!_116_!_news_game_!_王者农药，一场打到虚脱的游戏，大家最多打多长时间？_!_中路,塔残血,甄姬,发信号,正好接到朋友
+6552359891552436739_!_116_!_news_game_!_盘点那些年你一定玩过的的经典游戏，钻石恒久远，经典永流传！_!_GTA,红色警戒2,游戏,极品飞车,即时战略游戏,FPS游戏
+6527413895022248200_!_116_!_news_game_!_你为什么还在坚持玩《炉石传说》这个游戏？_!_
+6552382385525096968_!_116_!_news_game_!_DNF最强预言家：15年的朋友看过来！你绝对想不到的……_!_哥布林,DNF,代币券,勇士们,阿雷
+6552405517405257987_!_116_!_news_game_!_如何看待现在的UZI的水平？_!_
+6552376984897323278_!_116_!_news_game_!_2000左右买什么手机？_!_
+6552072640339640840_!_116_!_news_game_!_玩家求官方加强猴子，别改李白，策划一一答复_!_五五开,李白,王者荣耀
+6552363436120998151_!_101_!_news_culture_!_上联：老子骑牛读书，下联怎么对？_!_
+6552390950096732423_!_101_!_news_culture_!_赌石怎样选手镯料？_!_
+6552271838381080835_!_101_!_news_culture_!_古人出门携带的火折子，究竟是用什么做的？_!_
+6552456861050732803_!_101_!_news_culture_!_情牵悟空夜未眠，对下联吧？_!_
+6552255038264181006_!_101_!_news_culture_!_红脸红心红屁股，日日红运。如何接句？_!_
+6552383198062444813_!_101_!_news_culture_!_原创上联：白鹿原上无白鹿，下联如何对？_!_
+6552470089755525390_!_101_!_news_culture_!_上联：江湖河流海洋深，如何对下联？_!_
+6552414007964205316_!_101_!_news_culture_!_上联：秋风已满洛阳城，试问季鹰归未，下联怎么对？_!_
+6552293053019718152_!_101_!_news_culture_!_哇！太美了！这人毛笔字写得太漂亮了！真让人羡慕痴迷！_!_书法,毛笔,水写布,书写,毛笔画
+6552327888077586702_!_101_!_news_culture_!_上句：藕入泥中玉管通地理，下句怎么对好呢？_!_
+6552234163812958468_!_101_!_news_culture_!_上联，山高，水低，大地宽？_!_
+6552235822286897412_!_101_!_news_culture_!_孙二娘被武松羞辱调戏，为何张青不敢上去帮忙？_!_
+6552486990514749704_!_101_!_news_culture_!_上联：白石板桥通南北，下联怎么对？_!_
+6552338328232919300_!_101_!_news_culture_!_上联：若不撇开终是苦，如何对下联？_!_
+6552459817342992643_!_101_!_news_culture_!_上联：朝秦暮楚终有失；如何对下联？_!_
+6552389501648371981_!_101_!_news_culture_!_上联：宝鸡啄食陈仓米，怎么对下联？_!_
+6552490587591803396_!_101_!_news_culture_!_经纬织物，纵横承序_!_万载,夏布,织中之圣,传统工艺,工艺品
+6552453956063199502_!_101_!_news_culture_!_谁是人类真正的祖先，你有何依据？_!_
+6552129658316390670_!_101_!_news_culture_!_马克思是世界上最伟大的哲学家吗？_!_
+6552413642300588291_!_102_!_news_entertainment_!_当你开心的时候你会找谁分享？_!_
+6552432192037847559_!_102_!_news_entertainment_!_曹云金终于坦白，这么多年没有过师父，更没有师叔！_!_曹金云,曹云金,德云社,郭德纲
+6552433530180207112_!_102_!_news_entertainment_!_古装剧死的最“虐心”的配角，糖宝第六，最后一位才是真的痛心！_!_宇文邕,陆贞传奇,李君羡,韩晓东,锦绣未央,武媚娘,仙剑三,兰陵王,至尊红颜,杨雪舞,仙剑奇侠传一,花千骨,仙剑奇侠传三
+6552467574179758340_!_102_!_news_entertainment_!_有什么比较好看的电视剧可以推荐给我？_!_
+6552481983648236040_!_102_!_news_entertainment_!_黄家驹为什么被神话了？如何评价黄家驹的音乐水平_!_海阔天空,黄家驹,摇滚乐队,演唱会,邓紫棋
+6552302856244822536_!_102_!_news_entertainment_!_上映8天仅3216万，老戏骨再多也救不了，网友：没想到烂成这样_!_倪大红,战神纪,赤道,老戏骨,幕后玩家,低压槽：欲望之城,证人,扫毒,陀地驱魔人,盂兰神功,激战,苗侨伟,票房,张家辉
+6552412948113916419_!_102_!_news_entertainment_!_《花木兰》预算超18亿！刘亦菲成世上成本最高女性电影女主角！_!_花木兰,女性主角电影,长发公主,迪士尼公主,女主角,刘亦菲,迪士尼
+6552135344408494599_!_102_!_news_entertainment_!_北京女子图鉴：现实而残酷，这就是生活！_!_北京女子图鉴,北京,电影,陈可
+6552458382043775239_!_102_!_news_entertainment_!_如何评价演员周润发？_!_
+6552424901213422094_!_102_!_news_entertainment_!_大型纪录片《我到新疆去》首映，同名图书首发式成功在京举办_!_陈鲁豫,库尔班江,李秋平,我到新疆去,我从新疆来,磨铁图书,纪录片
+6552454223286501895_!_102_!_news_entertainment_!_岳云鹏10瓶辣酱换一张“原图”，陈赫：你有毒_!_搞笑艺人,陈赫,吴彦祖,岳云鹏
+6552449837667713550_!_102_!_news_entertainment_!_谢娜回归快本，何炅读张杰写给谢娜的信感动全场，维嘉哭成泪人！_!_快乐家族,快乐大本营,谢娜,张杰,好丈夫,何炅
+6552487267494002958_!_102_!_news_entertainment_!_央视主播朱讯近况如何？_!_
+6552339938610774532_!_102_!_news_entertainment_!_闵孝琳：被韩国网民评选为“最像人偶的女艺人”_!_Flapper,随风而逝,爱上浪漫，首尔,rinz闵孝琳,Triple,Touch me,阳光姐妹淘
+6552323929006932483_!_102_!_news_entertainment_!_佟丽娅短发一字肩低胸白裙，尽显优雅干练！_!_中央戏剧学院,宫锁心玉,母仪天下,佟丽娅,锡伯族,北京爱情故事,新不了情
+6552310016693502471_!_102_!_news_entertainment_!_漫威有嘻哈！3分钟RAP唱完《复仇者3》之前18部电影_!_黑豹,星际异攻队,漫威,RAP,无敌浩克,跪着听完,钢铁人,复仇者3,复仇者联盟3 ：无限之战,Rap,漫威电影宇宙
+6552347484524380685_!_102_!_news_entertainment_!_张基龙、秦基周主演新剧《过来抱抱我》！唯美海报公开_!_过来抱抱我,樱花树下,许峻豪,秦基周,水木剧,预告影片
+6552356265698263556_!_102_!_news_entertainment_!_郭晶晶婆婆嫁46亿富豪，如今60岁和赵雅芝同框，一身珠宝赢了_!_朱玲玲,罗康瑞,霍震霆,郭晶晶,赵雅芝
+6552374668001542664_!_102_!_news_entertainment_!_倪妮24岁才出道？看来只要努力，现在逐梦演艺圈也不晚！_!_改写命运,限量版,金陵十三钗,SK-II,蘑菇头,倪妮
+6552403737426526727_!_102_!_news_entertainment_!_八年之后，夏家三千金现在还好吗？_!_夏家三千金,唐嫣,刘亦菲,夏友善,夏天美
+6552288147131072772_!_103_!_news_sports_!_女乒姑娘夺冠后接受采访，盛赞日本姑娘在比赛中的发挥。这是怎么回事？_!_
+6552260541908255246_!_103_!_news_sports_!_「热点新闻」+24！塔图姆的正负值再次追平队史季后赛新秀单场_!_正负值,凯尔特人,艾尔-杰弗森,塔图姆,季后赛
+6552416703853101571_!_103_!_news_sports_!_6日体育精选：「人事观察」奥运冠军吴敏霞当选上海团市委副书记，这个岗位有多位体育明星_!_吴敏霞,瑞典队,霍福德,贝克汉姆,库里
+6552383054038434052_!_103_!_news_sports_!_科比他的一生之敌只有一个人，不是詹姆斯，那会是谁？_!_
+6552371516753838343_!_103_!_news_sports_!_为什么明明詹姆斯最后体力不支，但是投那个绝杀球的时候，还能跳那么高？_!_
+6552280310338289923_!_103_!_news_sports_!_世乒赛半决赛,小胖樊振东开启无敌模式，3:1暴击卡尔松。你怎么看？_!_
+6552368154851410435_!_103_!_news_sports_!_张玉宁再一次被排除出大名单，不莱梅保级了也不要他 今夏离队吧_!_西布朗,张玉宁,不莱梅,德甲,勒沃库森
+6552396845912424974_!_103_!_news_sports_!_小克里琴科都不是富里的对手，英国狂人却扬言将其KO出局_!_贝卢,维尔德,英国人,次重量级,小克
+6552432900963303688_!_103_!_news_sports_!_乔丹、科比、詹姆斯，谁才是季后赛的关键先生？_!_
+6552486153012904205_!_103_!_news_sports_!_格林和隆多在场上矛盾不断，为何双方教练却反过来称赞对方？_!_
+6552313324455854350_!_103_!_news_sports_!_詹姆斯再一次绝杀了，猛龙输球，你怎么看？_!_
+6552486527039963395_!_103_!_news_sports_!_如果利物浦在欧冠决赛中战胜皇马，萨拉赫就能击败梅西C罗稳拿金球奖了吗？_!_
+6552271436206047491_!_103_!_news_sports_!_5月6号，多伦多猛龙客场挑战克利夫兰骑士进行G3，猛龙可以获得胜利避免被骑士横扫吗？_!_
+6552482615645962766_!_103_!_news_sports_!_这锅太阳来背，球员在球场内发边线球没被发现_!_雷丁,足球,英冠联赛,英超,加迪夫城
+6552432309667103246_!_103_!_news_sports_!_图解｜2017年恒大俱乐部亏损近10亿，中超球队普遍超支_!_中超,足球,Sporting,普华永道,广州恒大淘宝足球俱乐部
+6552278250284909070_!_103_!_news_sports_!_足球彩票：弗格森爵士突发脑溢血入院_!_弗格森爵士,脑溢血,曼联,弗格森,足球彩票
+6552419214919991565_!_103_!_news_sports_!_山东鲁能客场0-0战平北京人和，失去登顶机会，怎么评价本场比赛？_!_
+6552276913056907790_!_103_!_news_sports_!_还是李晓霞会说话：啥也不说了东京拿下，今天男队也拿下齐活了_!_世乒赛,李晓霞,朱雨玲,刘诗雯,李隼
+6552495160893112580_!_104_!_news_finance_!_在股票市场当中，如何建立完整的趋势交易系统？_!_
+6552383666356486660_!_104_!_news_finance_!_轻型客车增幅最大！宇通前4月销售各类客车1.3万辆_!_宇通客车,中型客车,各类客车,客车,轻型客车,大型客车
+6552475387157283336_!_104_!_news_finance_!_卖气味竟卖出3个亿！她靠8万块创业成功，张艺谋、奔驰、杜蕾斯都是客户_!_气味图书馆,图书馆,娄楠石,气味,新西兰
+6552402540967756046_!_104_!_news_finance_!_为什么A股投资者越保护越亏？_!_
+6552334573454229768_!_104_!_news_finance_!_有什么项目可以投资，但不用自己盯着？_!_
+6552302192076784132_!_104_!_news_finance_!_二铵出口外高内低 后市或可期？_!_外高内低,东北地区,离岸价,成交价,二铵
+6552346471646102019_!_104_!_news_finance_!_下周大盘可能暴跌1000点_!_鹏翎股份,星星科技,高斯贝尔,填权,永泰能源,紫光国芯,二连板,智光电气,昊志机电,开板,永和智控,沙河股份,美丽生态,广联达,金刚玻璃,金运激光,江龙,华北制药
+6552437655844421902_!_104_!_news_finance_!_晚清的GDP世界第一，是日本的5倍，为什么还打不赢日本？_!_
+6552450570525868547_!_104_!_news_finance_!_国际油价降至两周最低，下一轮油价调整会下跌吗？_!_中国石油,垄断者,油价,原油期货价格,国际
+6552250467856941316_!_104_!_news_finance_!_余额宝升级后，有两只新的货币基金，分别是博时基金和中欧基金管理，对用户有什么风险吗？_!_
+6552262438610272771_!_104_!_news_finance_!_罚款逾1.8亿！银保监会开出首批银行大罚单，招行、兴业、浦发中招_!_金融机构,兴业银行,王兆星,银行业,浦发银行,中国银行,同业,理财产品
+6552300822225486344_!_104_!_news_finance_!_银行纷纷上调大额存单利率，部分银行较基准利率上浮达55%_!_银行业,国信证券,大额存单管理暂行办法,兴业银行,国泰君安,银行存款,大额存单,大额,股份制银行
+6552494459244773896_!_104_!_news_finance_!_20180506复盘：区间震荡等方向_!_趋势股,科技股,星星科技,沙河股份,金刚玻璃,建新股份,贸易战
+6552318715747107080_!_104_!_news_finance_!_利息与利息率是如何产生的？_!_
+6552478152348664077_!_104_!_news_finance_!_二十万用来投资期货好不好？_!_
+6552348864899187204_!_104_!_news_finance_!_0504复盘：市场情绪有待修复_!_贸易战
+6552497789387932173_!_104_!_news_finance_!_大爆料！著名黑嘴证券主持人，操纵股票被罚1.29亿元！_!_三友化工,证券法,石灰石,证监会,佳士科技,重点关注,廖英强,中兴通讯,每日经济新闻,MAC,中国移动,Facebook
+6552371443370295822_!_104_!_news_finance_!_国产软件+智慧城市第一龙头，涨停突破再创新高，能否一飞冲天_!_计算机硬件,进出口,广联达,电子计算机,外围设备
+6552490118790250766_!_104_!_news_finance_!_月薪12000怎么理财？_!_
+6552478000108012039_!_106_!_news_house_!_房贷利率迅速“六连跳”！为何还刹不住房价的上涨？_!_房价,炒房客,首套房,房贷利率上浮,房贷利率
+6552222429035364868_!_106_!_news_house_!_中原区高关注度楼盘多维度分析_!_陇海西路,工业园区,西四环,郑州市,郑州
+6552373430690251011_!_106_!_news_house_!_2018年青岛的房价走势情况如何？_!_
+6552370869673394691_!_106_!_news_house_!_「壹玖·案例」认筹1万元即可获得3500元的蔬菜一年后返还本金！_!_大米,蔬菜,思维方式,无息退还本金,回购模式
+6552460495809413389_!_106_!_news_house_!_一个城市到底要需要多少地铁？_!_
+6552332058524385805_!_106_!_news_house_!_无锡市锡东区《锡沪公馆》70年产区小户型，贰号线地铁口贰分钟_!_无锡东站,沪公馆,无锡,地铁口,小户型
+6552388776511930883_!_106_!_news_house_!_房产博弈，已至高点，你的房产还有救吗？_!_房价,房产,炒房者
+6552265175062282760_!_106_!_news_house_!_今天的等一等全款真的成为了首付……_!_房地产,青年节,吴晓波,凉凉,楼面价,起拍楼面价
+6552354852641440270_!_106_!_news_house_!_买房最激进！这个省，楼市将走向何方？_!_厦门,福建人,楼市,福建,住户贷款
+6552440983135453454_!_106_!_news_house_!_在西安每个月5000的工资能买房吗？_!_
+6552313433050579464_!_106_!_news_house_!_武汉：开发商不得将住房核查资料作为购房前置条件_!_开发企业,购房资格,房查,购房人提供,房产信息查询
+6552375311600713992_!_106_!_news_house_!_年轻人自己买房时，要注意什么？_!_
+6551661409699103235_!_106_!_news_house_!_户籍新政仍解决不了房源紧张，西安在售楼盘有哪些？_!_房源,高陵区,公积金,西安市,楼盘
+6552464957160554765_!_106_!_news_house_!_在房子贷款未还清的情况下，能否在房产证上增加他人名字？_!_
+6552354467122004483_!_106_!_news_house_!_知识科普｜在广元哪些人可以申请廉租房？_!_廉租住房保障,家庭成员,房地产权证,出示原件,交复印件,家庭人均
+6552420750186250500_!_106_!_news_house_!_大家租房一般是通过什么渠道找的？在青岛李沧区有什么小区可以推荐吗？_!_
+6552406536323334664_!_106_!_news_house_!_与菏泽 共未来 菏泽碧桂园媒体答谢晚宴圆满落幕_!_晚宴,菏泽碧桂园营销负责人贾吉印,碧桂园,菏泽电视台,碧桂园·天汇负责人谷传牧
+6552382695438024964_!_106_!_news_house_!_2018年，在鹰潭哪个区域买房好点？_!_
+6552399221130002696_!_106_!_news_house_!_你买房子时首付是怎么筹的？_!_
+6552458341732319491_!_107_!_news_car_!_雅阁，凯美瑞和天籁的后期费用，大约都是多少钱？_!_
+6552471344166994446_!_107_!_news_car_!_扭力梁后悬架、干式双离合变速箱，这样的柯珞克值得入手吗？_!_DQ200,柯珞克,斯柯达,双离合变速箱,SUV
+6552291228728164611_!_107_!_news_car_!_汽车在高速上超速百分之十以下会怎样处理？_!_
+6552264324113498628_!_107_!_news_car_!_网友：合资车低价的背后，为何是国产车越发拔高的价格？_!_合资车,自主品牌,高档车,国产车,ix35
+6552244747161829895_!_107_!_news_car_!_难怪老公把车子的脚垫都拆掉了，时下都玩这些，开车隔音还享受_!_中控台,密封条,汽车配件,扶手箱,储物盒
+6552339375928115719_!_107_!_news_car_!_奇瑞 致敬青春有梦的你！_!_奇瑞,艾瑞泽,艾瑞泽5,5x
+6552353200584802829_!_107_!_news_car_!_实体店很难买到的8款小玩意车品，瞬间提升爱车档次，高级实用！_!_车载吸尘器,护目镜,遮阳板,实体店
+6552297582909980932_!_107_!_news_car_!_帝豪GS帝豪GL选哪个好？_!_
+6552465018934264067_!_107_!_news_car_!_宝马530混动怎么样？_!_
+6552445436282536206_!_107_!_news_car_!_电动机轴承位磨损后用什么方法修复比较好？_!_
+6552272973363937549_!_107_!_news_car_!_新提车需要打腊，封釉或镀晶吗？_!_
+6552285328105800196_!_107_!_news_car_!_涡轮增压器要怎么样保养最好？_!_增压器,发动机,涡轮增压发动机,涡轮增压器,机油
+6552444283398390029_!_107_!_news_car_!_如何看待吉利为首的自主品牌已经超越韩系车了吗？_!_
+6552283921898275336_!_107_!_news_car_!_奔驰、宝马、路虎高级工程师进阶培训课程_!_Pathfinder,工程师,SDD,捷豹,路虎,揽胜,编程,NBT,DTS,宝马,CARPLAY,汽车维修,汽修学校
+6552385887861211652_!_107_!_news_car_!_新款Macan，我怎么觉得不如以前霸气呢_!_macan
+6551627130143244803_!_107_!_news_car_!_嘀，前方道路请掉头！_!_速度与激情,左转向灯,掉头
+6552245519635186183_!_107_!_news_car_!_车子保养别只知道换机油！这一“关键”部位不注意，油耗蹭蹭上涨_!_发动机,打气泵,保护剂,挡泥板,护目镜
+6552349411849011715_!_108_!_news_edu_!_昌乐北大公学双语学校用爱润泽学校教育，给孩子一个美好童年_!_昌乐北大公学双语学校,玥萱同学,爱的教育,昌乐,公学
+6552262326655910157_!_108_!_news_edu_!_杨振宁说：把中学生送到美国读书非常危险，你怎么评价？_!_
+6552444288221839880_!_108_!_news_edu_!_公务员面试无领导小组讨论_!_理解团队,调动组织成员,面试官,配合团队成员,领导小组讨论
+6552468719128281603_!_108_!_news_edu_!_《太原市2018年中考体育考试实施方案》正式发布_!_体育考试宣传手册,各县,太原市,考生家长通知书,体育考试流程及规则,立定跳远,残疾证,学校体育,考生成绩更正表,体育考试评分标准,考试成绩,体育考试,考生缓考申请表,体育考试操作流程,体育考试安全责任书,体育考试仲裁意见表,体育,申请
+6552277679490466307_!_108_!_news_edu_!_高考作文怎么考？先看看过去40年的作文题目再说……_!_作文题目,高考作文
+6552459713961787912_!_108_!_news_edu_!_大学生“心声”合唱受捧！_!_华南师范大学,苗族飞歌,沧海行,回家,雨霖铃•寒蝉凄切,心声·第四季,乐记,醉天山,合唱,音乐会,心声
+6552333073399152899_!_108_!_news_edu_!_重庆2018年文科生二模540左右，在全国内能上什么大学？_!_
+6552237858659238147_!_108_!_news_edu_!_软件工程专业就业前景？_!_
+6552306508049154573_!_108_!_news_edu_!_事业单位，公务员笔试面试干货，资料视频领取_!_行测,pan.baidu.com,面试,心竺公考,申论
+6552268250753794563_!_108_!_news_edu_!_滨州初三年级学业考试有新变化！语数外满分调整为150分_!_语文,北镇中学,招生,意见,滨州市,高中阶段,传统文化
+6552392806336299534_!_108_!_news_edu_!_北大校长林建华的校庆致辞是自己写的吗？_!_道歉信,校庆致辞,北大校长,林建华,林校长
+6552330480262316302_!_108_!_news_edu_!_大学挂科，考研会受到影响吗？_!_
+6552079480679039492_!_108_!_news_edu_!_5月12日北方工业大学将举行27所高校参加的联合咨询会和开放日_!_北京邮电大学,北方工业大学,北京工业大学,首都师范大学,　　
+6552483244070142216_!_108_!_news_edu_!_苏州大学全国排名第13位不太可能吗？_!_
+6552472654358512141_!_108_!_news_edu_!_弘扬“五四”精神，竞展青春风采_!_共青团,校团委,五四,团员青年,习总书记
+6552264127929123336_!_108_!_news_edu_!_小学生成绩不重要吗？_!_培养习惯
+6552273399626859015_!_108_!_news_edu_!_一场自己跟自己比赛的运动会——校长：越努力，越幸运！_!_校长,山东师范大学,平板电脑,运动会,体育运动
+6552405406017126659_!_108_!_news_edu_!_你对哪些世界未解之谜感兴趣？_!_
+6552396628995604999_!_108_!_news_edu_!_菏泽市第二届“慕清杯”国际足球邀请赛开赛_!_足球,菏泽市,单县,北京理工大学,邀请赛
+6552242693383127565_!_108_!_news_edu_!_为学生减压 南昌发布“十不准” 对校外培训说不_!_小学教师,青山湖区,十不准,有偿补课,校外培训机构
+6552319243700928775_!_109_!_news_tech_!_日本的芯片技术在世界上是什么水平？_!_
+6552407205151244552_!_109_!_news_tech_!_高佣联盟和淘宝联盟有什么区别？_!_
+6552399129786450180_!_109_!_news_tech_!_如果苹果放弃IOS采用安卓将会怎样？_!_
+6552327644120088840_!_109_!_news_tech_!_国内程序员那么多，为什么不开发一个属于自己的编程语言？_!_
+6552299275601052163_!_109_!_news_tech_!_雀巢和星巴克？_!_星巴克,西雅图,速溶咖啡,雀巢
+6552348138122773000_!_109_!_news_tech_!_华为手机销量上涨，余承东：即使没有美国市场，也要做世界第一！_!_手机市场,华为公司,西班牙,华为手机,市场份额
+6552312473091834115_!_109_!_news_tech_!_《唐人街探案2》中，女黑客在监狱里组装电脑，在现实中可行吗？_!_
+6552220450347286791_!_109_!_news_tech_!_谁有好的微信头像可以分享？_!_
+6552350940333081102_!_109_!_news_tech_!_央行发布新规定限制移动支付，将影响每个人，支付平台坏消息来了_!_接入网,支付宝,银行,余额宝,马云
+6552320624876847374_!_109_!_news_tech_!_安卓版本越高有什么好处？_!_
+6552368436968685828_!_109_!_news_tech_!_为什么自己做的有些交互，总是不尽如人意？_!_
+6552346504630108430_!_109_!_news_tech_!_哪些视频软件的录制效果好？_!_
+6552342412352553223_!_109_!_news_tech_!_现在最便宜的64G内存手机是1299元，未来10年，64G内存的手机能便宜到多少钱？_!_
+6552250609150460429_!_109_!_news_tech_!_欧洲5G已经启动，德国电信的5G天线已经启用；华为明年发布5G手机_!_运营商,天线,T-Mobile,智能手机,德国电信
+6552388824805146894_!_109_!_news_tech_!_你选过小众手机品牌吗，后悔吗？_!_
+6552230234127598083_!_109_!_news_tech_!_二手手机号遭遇解绑难题_!_运营商,支付宝,验证码,中国联通,中国电信,手机号码,刘兴亮
+6552386078475551245_!_109_!_news_tech_!_区块链带来颠覆性革命，大洋链Oeanc带来颠覆性认识_!_沃尔特,区块链,区块链技术,大洋链Oeanc,海洋贸易
+6552366936699699720_!_109_!_news_tech_!_京东天猫淘宝拼多多最简单最有效的刷单技巧_!_快递单号,淘宝刷单,天猫,流量入口,京东
+6552136956388573443_!_109_!_news_tech_!_联想被踢出恒生指数，是什么导致了联想现在的这种境地？_!_
+6492902897011917070_!_110_!_news_military_!_为什么军队不再使用三棱军刺？_!_毛诗.秦风.小戎,军队,周公东征,现代战争,三棱军刺
+6538627588812702215_!_110_!_news_military_!_毛主席天安门阅兵，高喊人民万岁，历史最伟大的时刻看一次哭一次_!_阅兵,毛主席,历史,天安门
+6548194293381595651_!_110_!_news_military_!_中国最窄的领土，一不留神就出国了！_!_珲春市,图门江,敬信镇,俄罗斯,图们江
+6546511919442821640_!_110_!_news_military_!_中国抗日战争，日军损失多少人？_!_日本,淞沪会战,靖国神社,整建制,日本军
+6546837092566040846_!_110_!_news_military_!_为什么电子狗会不允许使用？_!_
+6530675307152671235_!_110_!_news_military_!_一枚洲际导弹造价能有多贵？算完这笔帐你可别被吓到_!_洲际导弹
+6549863825460429316_!_110_!_news_military_!_400架废弃战机重新刷上漆就卖了1500亿，十几个国家排队求购_!_B-1,F-15,F-16,战斗机,预警机,反潜机,战机
+6552328700149367309_!_112_!_news_travel_!_印度人来中国旅游，被误认成巴基斯坦人，面对中国人的热情很尴尬_!_印度人,巴基斯坦人,巴基斯坦
+6552150431378702852_!_112_!_news_travel_!_天热别直接坐地上，一套干净凉爽的充气沙发床，户外不怕没地方睡_!_U型枕,沙发床,便携式,纸香皂,显微镜
+6552325824748454413_!_112_!_news_travel_!_实拍：缅甸另类的景区_!_
+6552319701937029645_!_112_!_news_travel_!_2018全球避暑名城排行榜出炉，六盘水入围百强！_!_六盘水市,贵州省,黑龙江省,山西省,青海省,避暑,辽宁省,六盘水,吉林省,云南
+6552335782386860548_!_112_!_news_travel_!_80%不需要门票！浙江100条旅游精品线路发布，周末就去浙里玩~_!_休闲农业,观光点,农业文化,美丽田园,乡村旅游
+6552486360161190403_!_112_!_news_travel_!_浙江湖州月亮湾：不用去迪拜，太湖上也有豪华七星级酒店_!_太湖,七星级酒店,湖州,月亮湾,渔人码头
+6552210229361115396_!_112_!_news_travel_!_一个50岁的单身女人，适合自驾游吗？_!_
+6552470876770533636_!_112_!_news_travel_!_夏天到了，红原月亮湾值得去吗？_!_
+6552278093334053390_!_112_!_news_travel_!_“十大最美茶乡”正在公示 宜宾三地入选_!_茶乡,南江县,平昌县,青川县,屏山县
+6552342474222731783_!_112_!_news_travel_!_绝地求生刺激战场沙漠地图：这两个又富又隐蔽的点你知道吗？_!_绝地求生：刺激战场,跳伞地点,SKS,沙漠
+6552444932508877064_!_112_!_news_travel_!_川藏到拉萨需要几天，每天行多少公里？_!_
+6552452075727356173_!_112_!_news_travel_!_去墾丁兩天游，需要包車帶我們遊覽嗎？_!_
+6552447230870028552_!_112_!_news_travel_!_国家为什么不允许用红砖建房？_!_
+6552421879250944264_!_112_!_news_travel_!_学厨师去哪里学好？_!_
+6552417256733671939_!_112_!_news_travel_!_希腊旅游撒狗粮_!_德尔菲,希腊,自由行,迈锡尼,哈尼亚,国家地理
+6552290479734522382_!_112_!_news_travel_!_微信史上最囧挑战15关图文攻略 史上最囧挑战15关五星好评答案_!_微信
+6551274851682746888_!_112_!_news_travel_!_初夏花开日，人生初见时，五月赏这几种花木正当时_!_油菜花
+6552420224111477005_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552489512369389837_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552128956781297924_!_113_!_news_world_!_越南以前叫什么？_!_
+6552149938988384519_!_113_!_news_world_!_美国法院判伊朗赔付9·11遇难者家属逾60亿美元，你怎么看？_!_
+6552296033626358280_!_113_!_news_world_!_三架俄战机冲向叙利亚基地？叙军解除锁定，结果炸弹落下引发地震_!_叙利亚政府军,苏35,战机,俄罗斯,F15,以色列,军事基地
+6552259946069623047_!_113_!_news_world_!_美国给联合国193名成员的表现打分，列“良好”及“不友好”清单，这是出于什么目的？_!_
+6552366471027098119_!_113_!_news_world_!_白冰冰送别母亲 悲叹母亲节将至妈妈却已不在_!_白冰冰,阿尔文
+6552468965447172355_!_113_!_news_world_!_如果现在是希拉里当总统，美国的外交政策会是什么样？_!_
+6552377851004322051_!_113_!_news_world_!_美国建国不过两百年，为什么会成为世界强国？_!_
+6552415736734679565_!_113_!_news_world_!_外交部：在华外企应尊重中国人民民族感情_!_中国外交部发言人,中国人民,中国领土,外交部,领土完整
+6552490895806038276_!_113_!_news_world_!_美国法院判伊朗赔付9·11遇难者家属逾60亿美元，你怎么看？_!_
+6552456966864650503_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552288209835917572_!_113_!_news_world_!_如果证实反政府武装使用化学武器，美英法会被追责吗？_!_
+6552294625757889027_!_113_!_news_world_!_见识下啥叫中国制作！美国一工程运行半天就瘫痪！最后还是靠中国_!_庞巴迪公司,加拿大,美国,蒙特利尔,洛杉矶
+6552136457593553165_!_113_!_news_world_!_美国建国不过两百年，为什么会成为世界强国？_!_
+6552310827184030222_!_113_!_news_world_!_强敌当前叙利亚反政府武装却爆发内讧，俄叙联军不禁偷偷乐了！_!_代尔祖尔省,叙利亚民主军,库尔德武装,代尔祖尔,杜鲁门
+6552460780715901197_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552270224828137732_!_113_!_news_world_!_如果证实反政府武装使用化学武器，美英法会被追责吗？_!_
+6552460661652193539_!_113_!_news_world_!_侣行在伊拉克，叙利亚等的土豪行为，为后来的探险家造成巨大的伤害，您怎么看？_!_
+6552368074656317965_!_113_!_news_world_!_你的钻戒要贬值了！英国实验室仅花费数天成功制作一克拉人工钻石_!_实验室,人工钻石,钻石,碳原子,天然钻石
+6552025963360682509_!_115_!_news_agriculture_!_愁死了，水稻虫害不认识？专家帮你指认_!_二化螟,悬浮剂,微乳剂,幼虫,毒死蜱
+6552234456088838664_!_115_!_news_agriculture_!_你什么时候回来，不要在大城市打工了_!_嫁祸于人,农村
+6552416190868750606_!_115_!_news_agriculture_!_鸡农药中毒后，有什么好的解救方法？_!_
+6552234887284261383_!_115_!_news_agriculture_!_农村流传的“婆媳关系”顺口溜，句句有理！婆媳怎能是天敌！_!_顺口溜,婆媳关系,婆媳,农村
+6552120632165794317_!_115_!_news_agriculture_!_未来七天，农村2样东西降价，这2样东西大涨，农民要提前做好准备_!_水泥厂,农民朋友,农村,环保税征收,样东西
+6552445098133553411_!_115_!_news_agriculture_!_怎样化解洗车店位置不是很好的尴尬局面？_!_
+6552449646524891652_!_115_!_news_agriculture_!_湖北宜昌：乡村，农家楼房造型奇葩！这是什么样的建筑风格？_!_农家,宜昌市,湖北,岩花村,建筑风格
+6552407575537648141_!_115_!_news_agriculture_!_皖南黑桃树，产量极高，果香浓郁，当年结果，是庭院种果树皇后！_!_蜂蜜,皖南,皖南黑桃,果香
+6552487254089007364_!_115_!_news_agriculture_!_我想回家养牛，打算养10头左右，有什么需要注意的吗？_!_
+6552437589624750350_!_115_!_news_agriculture_!_孩子在城市养和在农村养有什么区别？_!_
+6552354049625161998_!_115_!_news_agriculture_!_什么树可以生长灵芝？_!_
+6552301665964261892_!_115_!_news_agriculture_!_农民工工资是否已经超过大学生？_!_全国农民工,2017年农民工监测调查报告,大学生
+6552314398076043789_!_115_!_news_agriculture_!_老照片：那时候吃的清贫，却热热闹闹，看看中国农村吃饭的情景_!_老照片,农村
+6552268555033772557_!_115_!_news_agriculture_!_我敢说这种美食零零后肯定没吃过，吃过的也是上了年纪的人_!_冷蒸,麦蚕,农村
+6552412321853997581_!_115_!_news_agriculture_!_在农村栽种的一种菜名叫牛皮菜，它是这样吃的_!_东方IC,农村,莙荙菜,牛皮菜,厚皮菜
+6552267237049238024_!_115_!_news_agriculture_!_农村野外做野味，用一口大锅做，场面太壮观了！_!_野猪,大锅菜,农村
+6552305507573432580_!_115_!_news_agriculture_!_对于蒜农来说，什么时候卖蒜合适？_!_
+6552464147882180867_!_115_!_news_agriculture_!_如何能有效，快速的提高我们的专业知识（畜牧类）？_!_
+6552393876307444232_!_115_!_news_agriculture_!_齐天大圣的“金箍棒”可以吃哦_!_齐天大圣,农村
+6552361065391325444_!_116_!_news_game_!_CF武器9A91怎么样？_!_
+6552483172414652686_!_116_!_news_game_!_《王者荣耀》新英雄米莱狄将在5月上线，她的出装攻略是怎样的？_!_
+6552127246193132045_!_116_!_news_game_!_游戏设备大升级，助你畅快吃鸡_!_掠夺者,笔记本,MacBook Pro,宇宙飞船,固态硬盘,独立显卡,绝地求生,Razer
+6552487109951750664_!_116_!_news_game_!_斗鱼嘉年华美年达展会艳武汉，人气爆棚成为最大赢家！_!_第五人格,美年达,王者荣耀,战区,格斗游戏,斗鱼,嘉年华
+6552406515486032131_!_116_!_news_game_!_如果自身四维，武力，智力，社交能力，工作能力，按游戏数据各满分100来算，你是多少？_!_
+6552461696277938439_!_116_!_news_game_!_大神飞车手游里结过婚了，戒指可以换吗？_!_
+6552299308496978184_!_116_!_news_game_!_DNF攒蓝矿+红豆换荒古是一种什么样的体验？_!_
+6552456973558743555_!_116_!_news_game_!_天美的五个亲儿子，皮肤很多很炫酷，可惜当前版本完全被抛弃！_!_韩信,王者荣耀,亲儿子,赵云,李白
+6552254984820359428_!_116_!_news_game_!_我的世界，奶块，迷你世界，泰拉瑞亚，麦块哪个更好？_!_
+6552293959534641678_!_116_!_news_game_!_梦幻西游：土豪被美女骗钱骂屌丝，一怒之下买175张号复仇_!_梦幻西游,游戏
+6552387411064652292_!_116_!_news_game_!_DNF：每日深渊任务完成，1000票回票500_!_dnf,深渊
+6552428671481676036_!_116_!_news_game_!_神武3天罡什么队伍最好？_!_
+6551695635349766403_!_116_!_news_game_!_在《王者荣耀》中玩射手应该注意些什么？_!_
+6552491388007612935_!_116_!_news_game_!_王者荣耀：后期最强势的4个法师，最后一个让敌人头疼，无解秒人_!_王者荣耀,干将莫邪,游戏,法师
+6552256899121676547_!_116_!_news_game_!_QQ飞车还火吗？_!_
+6552496549073518851_!_116_!_news_game_!_DNF会“后跳X”的职业都有哪些？_!_
+6552467849061859844_!_116_!_news_game_!_《大话西游2》回忆游戏，经历人生 大话回忆录，回忆你我她_!_斧头帮,玩家,华山论剑,北俱芦洲,人族,大话西游
+6552456332526486020_!_116_!_news_game_!_DNF：十亿小目标第四十四天_!_搬砖,dnf,游戏
+6552323831766188552_!_116_!_news_game_!_梦幻西游：玩家鉴定150装备时喜出破血，不料却只能当陪衬品！_!_梦幻西游,梦幻西游老板,梦幻西游玩家
+6552240543185764616_!_116_!_news_game_!_怎么看微博大V爆料王者掉帧卡顿等问题是因为手机品牌保护费交的不够？_!_
+6552310725006590215_!_101_!_news_culture_!_上联：西江月下西江美，如何对下联？_!_
+6552267798981116163_!_101_!_news_culture_!_上联：去过成都，都成过去。求下联？_!_
+6552437309013229831_!_101_!_news_culture_!_上联：夏隐云天哭春姐，如何对下联？_!_
+6552464476489122318_!_101_!_news_culture_!_诗词与联：「七律为遣春愁」闻莺懒唱花篱外；请对下联_!_七律.为遣春愁,七律
+6552459733427552520_!_101_!_news_culture_!_上联：梅花喜欢漫天雪，如何对下联？_!_
+6552227444487618819_!_101_!_news_culture_!_上联：一季花香游人醉，怎么对下联？_!_
+6552286242866725384_!_101_!_news_culture_!_书画素材｜唐代《大达法师玄秘塔碑》裴休撰文，柳公权书并篆额_!_大法师,虚舟题跋,艺概,柳公权,玄秘塔,郭家庙,大达法师玄秘塔碑,玄秘塔碑,安国寺
+6552385740641141006_!_101_!_news_culture_!_为何鲁迅写的“我家门前有两棵树，一棵是枣树另一棵也是枣树”成为名句，而不是病句呢？_!_
+6552446469020844292_!_101_!_news_culture_!_上联：风花雪月天上客，下联怎么对？_!_
+6552452699936260355_!_101_!_news_culture_!_为啥一根破烂的金丝楠木能值2000万？_!_
+6552303475697385736_!_101_!_news_culture_!_出联：落英缤纷凌空扬，谁知何处思故乡，如何接联？_!_
+6552323556812783880_!_101_!_news_culture_!_上联：枝上柳棉吹又少，天涯何处无芳草。下联是什么？_!_
+6552232786634211591_!_101_!_news_culture_!_古代没有电，天黑了他们都在干什么？_!_
+6552265375105417480_!_101_!_news_culture_!_上联：风花雪月天上客，下联怎么对？_!_
+6552366906194526727_!_101_!_news_culture_!_有听损的人怎样去听音乐_!_贝多芬,班杰明,交响曲,音乐家,听音乐
+6552256887688003843_!_101_!_news_culture_!_上联：杯中有酒赏明月，怎么对下联？_!_
+6552455014751666445_!_101_!_news_culture_!_上联：心里有座山，赵本山，怎么对下联？_!_
+6552393686469050887_!_101_!_news_culture_!_书荒部落 100本经典网络小说排行榜——2018.5.6_!_恋物癖,长白山脉,陈二狗,黑瞎子,异装癖
+6552321628494102798_!_101_!_news_culture_!_书房一般挂什么字画？_!_
+6552314853023809799_!_101_!_news_culture_!_上联：桃李春风一杯酒，如何对下联？_!_
+6552437998284177678_!_102_!_news_entertainment_!_你希望赵丽颖接什么样的剧或演什么样的角色？_!_
+6552477615557444110_!_102_!_news_entertainment_!_陈亦飞回应病情：抱歉让大家担心了，并附上几张与谭维维的自拍_!_谭维维,陈亦飞
+6552444408757748238_!_102_!_news_entertainment_!_谢娜高调C位回归大本营，四件事可见她在圈中的地位不简单_!_腊像馆,谢娜,我是大侦探,吉尼斯世界纪录,奚梦瑶,快乐大本营,杜莎夫人
+6552476875816436238_!_102_!_news_entertainment_!_谢娜微博粉丝破亿，累计获3项吉尼斯世界纪录，你怎么看？_!_微博,谢娜,蔡徐坤,吉尼斯世界纪录,炸子鸡,武林外传,快乐大本营
+6552436298433430029_!_102_!_news_entertainment_!_《爱情公寓》“小姨妈”竟和赵丽颖飙过戏，颖宝被花式扇脸太心疼_!_南宫燕,爱情公寓,十全九美,龙套史,陆贞传奇,宫2,唐悠悠,赵丽颖,邓家佳,楚乔传,锁清秋,花千骨
+6552455218565480973_!_102_!_news_entertainment_!_崔泰俊与好兄弟聚会 女友朴信惠西班牙独乐_!_池昌旭,朴信惠,崔泰俊,李宗泫,CNBLUE
+6552464135748059661_!_102_!_news_entertainment_!_王晶与他绝交，蔡卓妍与他撇清关系，网友：他跟猥琐男一样不讲理_!_杜汶泽,言论自由,大本营,蔡卓妍,黄秋生
+6552359763663913479_!_102_!_news_entertainment_!_女星阿什丽·格林尼出行，累了就很随意坐在街边休息_!_街边休息女星阿什丽,女星,丽·格林尼出行,阿什丽·格林
+6552461033032647176_!_102_!_news_entertainment_!_唐嫣被逼问什么时候与罗晋结婚，她的回答被赞高智商_!_高智商,杨幂,归去来,粉丝,罗晋,唐嫣
+6552489158768591364_!_102_!_news_entertainment_!_Sunnee“C位被消失”，爆经纪公司惹火主办方，镜头遭删减_!_创造101,Sunnee,全网黑,李子璇,强东玥,吴宣仪
+6552317433103778051_!_102_!_news_entertainment_!_如何评价防弹少年团金南俊的颜值？_!_
+6552407537231069700_!_102_!_news_entertainment_!_180506 一件背心都能撩妹无数 吴亦凡穿背心了解一下？_!_羽绒服,吴亦凡,背心,衬衫
+6552449294291436045_!_102_!_news_entertainment_!_杨幂着黑色抹胸裙优雅轻曼 微微浅笑温婉动人 戒指却戴在了食指上_!_杨幂,国家奥林匹克体育中心,闭幕式,电影节,北京大学生电影节
+6552462284642320910_!_102_!_news_entertainment_!_闫妮戴眼镜穿牛仔衣搭配运动裤现身机场，戴着口罩也能看到笑容！_!_运动裤,闫妮,牛仔衣搭配运动裤现身机场,东方IC,戴眼镜口罩遮面穿牛仔衣
+6552414672895607309_!_102_!_news_entertainment_!_《复仇者联盟4》2019年将带来漫威最黑暗的电影！_!_漫威,只想告诉你,克里斯托弗,复仇者联盟3 ：无限战争,复仇者联盟4,麦克菲
+6552309062258655501_!_102_!_news_entertainment_!_烤猪蹄的佐料有哪些？_!_
+6552331835790066184_!_102_!_news_entertainment_!_看了张馨予穿深V裙，再看看范冰冰，真想对李晨说一句：好福气！_!_李晨,山水画,凤囚凰,张馨予,范冰冰,v裙
+6552300818152817165_!_102_!_news_entertainment_!_穿山寨衣演激情戏，“台湾第一气质美女”怎么了_!_张钧甯,刘亦菲,毛晓彤,激情戏,温暖的弦,知识产权
+6552274881440907784_!_102_!_news_entertainment_!_温暖的弦大结局：朱临路放手温暖 去澳门找到一个喜欢的姑娘_!_张钧甯,占南弦,朱临路,张嘉倪,大结局
+6552360551555531022_!_102_!_news_entertainment_!_《创造101》首次小组对抗，吴宣仪的小组为什么又一次输给了sunnee那一组？_!_
+6552278641793827342_!_103_!_news_sports_!_2018“丝绸之路·美丽大荔”渭南国际马拉松赛鸣枪开跑_!_大马,大荔县,马拉松,大荔,渭南
+6552385227103142151_!_103_!_news_sports_!_如何评价TT和卡戴珊餐厅用餐引起克城球迷恐慌的事情？_!_
+6552461387166122244_!_103_!_news_sports_!_世乒赛夺冠后，李晓霞霸气送上祝福：“赢中国队？你们得等”，对此你怎么看？_!_
+6552474885606605325_!_103_!_news_sports_!_什么造就了今年的季后赛隆多_!_常规赛,季后赛,尼克拉,隆多,米罗蒂奇
+6552244549580751111_!_103_!_news_sports_!_为什么普遍认为不管怎么交易，东部其他球队还是打不过骑士？_!_
+6552149743265382663_!_103_!_news_sports_!_如何看领头羊上港的主场，终被国安2:1攻破？_!_
+6552448666634813960_!_103_!_news_sports_!_詹姆斯又双叕绝杀了！我们要停止勒布朗不能进关键球的言论！_!_猛龙队,勒布朗詹姆斯,森林狼,骑士队,詹姆斯
+6552402128722198797_!_103_!_news_sports_!_詹姆斯天神下凡压哨绝杀猛龙，德罗赞犹豫的眼神坐在替补席上不知所措，你如何评价？_!_
+6552307343143469315_!_103_!_news_sports_!_足球比赛中守门员什么情况下可以用手，什么时候又不能？_!_
+6552147263697715719_!_103_!_news_sports_!_太有才！赛后记者见面会，朱雨玲三言两语，让全场人员哈哈大笑_!_日本队,朱雨玲,刘诗雯,中国队,平野美宇
+6552465020817506830_!_103_!_news_sports_!_乐福复苏，骑士见招拆招，詹姆斯绝杀，猛龙被逼入绝境_!_洛瑞,猛龙,詹姆斯,乐福,骑士
+6552470984006304007_!_103_!_news_sports_!_有人说游泳比跑步的减肥效果更好，是真的吗？为什么？_!_
+6552301032095875591_!_103_!_news_sports_!_周日中超盘口分析_!_中超,鲁能泰山,天津泰达,山东鲁能,山东鲁能泰山
+6552411992798265607_!_103_!_news_sports_!_斯诺克世锦赛结束后，小丁世界排名会掉到多少位？_!_
+6552471348248052231_!_103_!_news_sports_!_彪马Puma Fenty Creepe_!_彪马,Puma,Fenty,SIZE,ermuptwx
+6552449401346851341_!_103_!_news_sports_!_就今天的情况来看，到底是低估了骑士队，还是高估了猛龙队呢？_!_克利夫兰骑士队,科沃尔,骑士队,詹姆斯,猛龙队
+6552366734441971972_!_103_!_news_sports_!_鲁能“劳模”强势复苏，李霄鹏被征服，里皮是否心动？_!_
+6552320536473501959_!_103_!_news_sports_!_猛龙被认为是打出队史最佳，为何就是打不过勒布朗？_!_
+6552270047094505991_!_103_!_news_sports_!_世乒赛1/4决赛有意外也有惊喜！小胖希望能排到第1位出场！_!_奥地利队,世乒赛,奥地利,樊振东,马龙
+6552275559395623431_!_103_!_news_sports_!_郜林：一方状态信心比我们强 盼亚冠发挥真正的水平_!_亚冠比赛,中超联赛,盼亚冠,郜林,广州恒大
+6552128267824922883_!_104_!_news_finance_!_市场比较认可两本理论，《波浪理论》《缠中说缠》大家对这两本理论有什么深入的见解？_!_
+6552371210238296589_!_104_!_news_finance_!_沙特上调油品价格，全国人民喜迎油价上涨_!_OPEC,欧佩克,油价,石油,原油,中石化
+6552440807889043981_!_104_!_news_finance_!_经历追涨杀跌风雨，终见价值投资彩虹？_!_A股市场,三一重工,投资者,价值投资,国债期货,四川长虹
+6552328711184581128_!_104_!_news_finance_!_秦绪文：全民创业时代到底什么才是最重要的？_!_互联网思维,秦绪文,互联网,经营者,创业者
+6552372211649347848_!_104_!_news_finance_!_如果散户都不交易了，股市会怎么样？_!_
+6552314867737428238_!_104_!_news_finance_!_几个朋友合伙创业，如何分配股权？_!_
+6552323639281189123_!_104_!_news_finance_!_钢价大涨后，会掉头大跌吗？_!_
+6552301696268108301_!_104_!_news_finance_!_扬中创业大讲堂精彩开讲_!_扬中市,创业者,大学生创业园,阶段—,国际劳工组织
+6552256148479672589_!_104_!_news_finance_!_如何判断股市突破是否有效？_!_
+6552399921830428941_!_104_!_news_finance_!_美元加息的含义和目的是什么？_!_
+6552413479343489544_!_104_!_news_finance_!_看看这些吸血虫的微信，就知道空贷套路深了！_!_吸血虫,借款人,空贷
+6551990590504960526_!_104_!_news_finance_!_一位老股民的血泪史：中国股市永远也看不厌的一篇股市文章！_!_QFII,成交量,投资组合,空头市场,工业指数,生物医药,股价,跟庄,主力,巴菲特,可口可乐公司,股票市场,庄家,可口可乐,二级市场
+6551631817378628109_!_104_!_news_finance_!_清华女教授曝：看懂成交量选股技巧，三万赚三千万，建议股民收藏_!_成交量,沙河股份,投资者,股价上涨,原动力
+6552280056868110852_!_104_!_news_finance_!_深圳文博会大力提升专业化办展水平_!_专业化,采购商,知识产权,文博会服务质量标准化手册,文博会,参展商
+6552300356435444237_!_104_!_news_finance_!_通过并购把公司做大，这个华人的影响力比马云、李嘉诚还大_!_半导体,安华高,博通,博通公司,李嘉诚,陈福阳
+6552266489938837773_!_104_!_news_finance_!_怎样做白酒？_!_
+6552340596319584515_!_104_!_news_finance_!_什么现象是金融危机爆发的前兆？_!_
+6552445548815712515_!_104_!_news_finance_!_股票除了融资融券，场外配资怎么筛选正规公司？_!_
+6552378077865836814_!_104_!_news_finance_!_为什么美联储不属于美国政府？_!_
+6552302538471768583_!_104_!_news_finance_!_「滚雪球」今天你戴“周大生”了吗？_!_老凤祥,珠宝首饰,加盟店,爱迪尔,周大生,加盟费,周生生,周大福
+6552408705155990029_!_106_!_news_house_!_未来房价是涨还是跌？楼市如此严峻调控之下，你该怎么办_!_房价,深圳市,任志强,通货膨胀,阿里巴巴
+6552296897384546573_!_106_!_news_house_!_房产证上没有男方名字，男方的私生子能继承房产吗？_!_
+6552466693971509517_!_106_!_news_house_!_湘潭有哪些建筑特色？_!_
+6552425024316244231_!_106_!_news_house_!_手中资金大约600万闲置资金，房产已限购，该怎么配置投资？_!_
+6552283649918632462_!_106_!_news_house_!_《六大任》——预测式样的断解方法的五个方面！_!_主要矛盾,四爻之中,驿马星煞,下克上,大六壬金口诀
+6552402767678276109_!_106_!_news_house_!_敲黑板啦！毛坯房验房收房攻略合集，同学们快来看！_!_座便器,电源插座,耐水腻子,验房,验房师
+6552402638011367688_!_106_!_news_house_!_十年后房子真的降到普通人可以接受的二三十万一套吗？_!_
+6552479952002875661_!_106_!_news_house_!_任志强关于的房价降价的预测，会在今年变成现实吗？_!_
+6552300434231394830_!_106_!_news_house_!_2018，看看全国有哪些地方装配式建筑规划目标出台了？（上篇）_!_装配式建筑评价标准,建筑产业现代化,建筑面积,装配式建筑,办公厅,政府投资
+6552398345061532168_!_106_!_news_house_!_独家发布｜2017中国房企明星CFO五十强_!_碧桂园,房企,负债率,管理能力,融创中国
+6552438510412890371_!_106_!_news_house_!_如果2018年房价趋稳，2019年是否会赢来爆发？_!_
+6552275731982844424_!_106_!_news_house_!_前四月销售榜单出炉 五家房企突破千亿_!_房企,碧桂园,万科,融创,上海院子
+6552465418420748548_!_106_!_news_house_!_四线城市的商铺，开发商卖170万，目前租金最多7万，可以入手吗？_!_
+6552283921856332295_!_106_!_news_house_!_租赁合同未到期但业主要卖房怎么办_!_违约金,出租人,当事人,租赁期限,房屋使用权,合同法
+6552392249588580612_!_106_!_news_house_!_能不能通过众筹自建楼拉低房价，比如有100户人家需要房，每户出资30万建好再分？_!_
+6552248601425215747_!_106_!_news_house_!_农村人在城里买房，到底值不值得？_!_
+6552402612182843656_!_106_!_news_house_!_海南买房先签订购房合同，挂靠公司交两年社保，满两年后再过户，这样做靠谱吗？_!_
+6552463879299924228_!_106_!_news_house_!_在浙江温州滨海碧桂园有一套89平的房子，94万可以买吗？_!_
+6552482803378815240_!_106_!_news_house_!_如果高层拆了，地底下的桩怎么办？_!_
+6552321836372197645_!_107_!_news_car_!_为什么有些开好车的女生很漂亮？_!_
+6552333247546654990_!_107_!_news_car_!_飞机主起落架是怎么刹车的？_!_
+6552280311680467469_!_107_!_news_car_!_厉害了，我的老婆，偷偷在车里装了这些神器，低配车档次直逼顶配_!_MP3播放器,        ,打蜡机,吸尘器,     
+6552354201739985160_!_107_!_news_car_!_大众速腾1.2T涡轮增压和1.6自然吸气自动挡，哪个好？_!_
+6552301974761505283_!_107_!_news_car_!_国内最牛的3快车牌，车牌不是连号，但连劳斯莱斯都不敢惹！_!_A6L,劳斯莱斯,车牌,奥迪A6L,奥迪a6l
+6552286137740689667_!_107_!_news_car_!_开手动挡车子，在做降档补油的时候，在踩下离合的同时补一脚油，挂档需要松油门么？_!_
+6552459447870947853_!_107_!_news_car_!_汽车市场最热门的SUV，20万以内它们必定上榜！_!_小型SUV,昂科拉,ix35,科雷嘉,逍客
+6552485750682681859_!_107_!_news_car_!_红旗H5上海正式上市，“老法师”也挡不住它的诱惑！_!_BOSE,红旗H5,红旗H7,B级车,老法师,BLIS,红旗
+6552460595981976072_!_107_!_news_car_!_这些中国力量纷纷崛起，厉害了我的国！_!_新能源,逸动XT,长安CS75,BJ80,长安汽车,科技感,比亚迪,Concept,BJ40,北京汽车,长安,CS75,比亚迪唐,新车,红旗
+6552172852408222222_!_107_!_news_car_!_新款上市“360鸟瞰”行车记录仪，一体多用，前后双录，忒实用_!_ADAS,防碰瓷,汽车镀膜剂,电源车,行车记录仪
+6552292924959228419_!_107_!_news_car_!_苏E88888等红灯，周围车里的人都在拍，当地的明星车_!_车牌,越野车,JEEP,苏州,CCTV
+6552085311608848900_!_107_!_news_car_!_凯美瑞的终极对手来了，轴距2.79米，配12.3英寸大屏，或售16万！_!_凯美瑞,豪车,标致,雅阁,标致508
+6552294423848288525_!_107_!_news_car_!_2.0奇峻两驱舒适和科雷傲2.0两驱豪华哪个更值得买？_!_
+6552404010358276612_!_107_!_news_car_!_4月SUV投诉排行：欧蓝德首次上榜_!_CS10,广汽三菱,宗典型故障,本田XR-V,长安汽车,CR-V,昂科威,东风本田,机油,CS75,GS4,欧蓝德,哈弗,本田CR-V,质量投诉点评,传祺GS4,长安CS75,XR-V,猎豹CS10,广汽传祺
+6552382961105240323_!_107_!_news_car_!_新手司机怎样才能不紧张？_!_
+6552375996878684680_!_107_!_news_car_!_2018北京车展：比亚迪宋MAX插混版正式亮相_!_北京车展,MAX,比亚迪,新车
+6552431133814948359_!_107_!_news_car_!_真正可以碾压特斯拉models的车，充电4分钟能跑100KM，来自保时捷_!_科技感,特斯拉,Panamera,Mission E,保时捷,碾压特斯拉
+6552277212261777934_!_107_!_news_car_!_5月27日上市 SENIA R9明日正式下线_!_R18,R17,SENIA R9,速双离合,森雅R7
+6552174256539566605_!_108_!_news_edu_!_景贤招生范围新增紫茶小学北校区 哪些楼盘将受益？_!_蓬江区,特长生,东方雅居,紫茶,江海区
+6552288258485649933_!_108_!_news_edu_!_艺术设计专业申请要点_!_美国院校,语言成绩,艺术设计专业,艺术设计,美国
+6552328133133992199_!_108_!_news_edu_!_考试，县级事业单位和公务员，应该选哪个？_!_
+6552414424303403268_!_108_!_news_edu_!_加拿大留学生移民后的生活状况如何？_!_
+6552415685425758478_!_108_!_news_edu_!_专科临床医学毕业，可以考什么证？_!_
+6552494991183184391_!_108_!_news_edu_!_雄安大事记：雄安新区招聘650人，还给住房补贴！_!_教师资格证书,住房补贴,雄县,中小学,幼儿园
+6552488597629436164_!_108_!_news_edu_!_葡萄树怎样才能挂果更多？_!_
+6552276738087322125_!_108_!_news_edu_!_市民政局团市委联合办班培训社工_!_湛江市,广东海洋大学,社会工作,社会工作者,团市委
+6552391628399575304_!_108_!_news_edu_!_怎样看待成都一高校让学生集体种草莓，还要写种植报告这件事？_!_
+6552468015202435335_!_108_!_news_edu_!_女教师讲台倒下昏迷30天，醒来问"我们班学生谁管"，你怎么看？_!_
+6552384437286339080_!_108_!_news_edu_!_今年兵团计划招聘4017名事业单位工作人员_!_事业单位,综合应用能力,新疆,兵团,职业能力倾向测验,教师
+6552208320034242824_!_108_!_news_edu_!_和同桌发生过哪些难忘的事？_!_
+6551649041443193352_!_108_!_news_edu_!_2018教师招聘考试｜心理学简述题简答题48道！_!_心理学,教师,简答题
+6552259683069985288_!_108_!_news_edu_!_河南大学终于在河南人民的期盼中进入了“双一流”，不容易啊！_!_南洋公学,郑州大学,河南,河南大学,张镇芳
+6552409201694474499_!_108_!_news_edu_!_孩子学习成绩全班倒数，为什么老师不能抽时间为孩子补补课，老师不应该帮孩子提高成绩吗？_!_
+6552476681578217997_!_108_!_news_edu_!_中考语文阅读题的解题和答题诀窍：关于记叙文、议论文和说明文_!_××,议论文,记叙文,说明,事物,散文
+6551640843659248136_!_108_!_news_edu_!_2018高考冲刺：权威名师详解物理、化学得分必杀技_!_北京新东方学校,名师,新东方学校
+6552444759212818691_!_108_!_news_edu_!_新乡医学院临床医学专业录取，这个学校怎么样啊？_!_
+6552268368726983172_!_109_!_news_tech_!_突然！腾讯王卡拉黑阿里！_!_阿里小号,阿里,亲情号码,腾讯
+6552380373471330830_!_109_!_news_tech_!_联想物流如何拥抱新零售_!_智慧零售,SEC,零售业,线上线下,物流体系,联想
+6552254181304959236_!_109_!_news_tech_!_为什么国内用户不怎么使用iOS自带应用？_!_
+6552490400622313741_!_109_!_news_tech_!_霍金给人类留下了几个预言？_!_
+6552353470010097927_!_109_!_news_tech_!_小米7会使用MIUI10吗？_!_
+6552309525066547459_!_109_!_news_tech_!_格力和小米赌约还算数吗？_!_
+6552272957329113357_!_109_!_news_tech_!_如何制作一个简单的手机app？_!_
+6552472460669747469_!_109_!_news_tech_!_经济学家吴敬琏为什么反对“不惜代价发展芯片产业”？_!_
+6551986279133741572_!_109_!_news_tech_!_阿里智能办公3.0，高级系统提升效率，在阿里办公也是一种福分！_!_阿里云,阿里,马云,人工智能
+6552295819393892877_!_109_!_news_tech_!_店大欺客，我们都是被互联网企业欺负的普通人_!_互联网企业,携程,店大欺客,滴滴打车,饿了么
+6552127687920451847_!_109_!_news_tech_!_传统PC互联网页正遭到边缘化，PC真的“已死”吗？_!_
+6552204337152524548_!_109_!_news_tech_!_有没有什么可以挣零花钱的app？_!_
+6552226705501585677_!_109_!_news_tech_!_如下图，电脑开不起来了，重新ghost恢复备份的系统后启动依旧蓝屏，怎么办？_!_
+6552307583833604359_!_109_!_news_tech_!_传魅族将裁员千人，裁员的原因是什么？_!_
+6552207584227492104_!_109_!_news_tech_!_如下图，电脑开不起来了，重新ghost恢复备份的系统后启动依旧蓝屏，怎么办？_!_
+6552270287356821767_!_109_!_news_tech_!_一个浙江省保险代理公司，如何利用互联网开展业务？_!_
+6552396144746430979_!_109_!_news_tech_!_腾讯没有梦想？马化腾：理想是做出最好的产品，而不是赚多少钱_!_腾讯,腾讯没有梦想,马化腾
+6552278067769770253_!_109_!_news_tech_!_创业者如何写出一份打动投资人并能首次融资3000万的商业计划书？_!_
+6552295617941471495_!_109_!_news_tech_!_如果所有银行不让支付宝绑定银行卡了，支付宝会怎样？_!_
+6552351006728913166_!_109_!_news_tech_!_没有电脑但有网线和路由器，电视机怎样才能看到网络电视呢？_!_
+6552419401218392323_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552231006936498445_!_110_!_news_military_!_当年德国为何放弃英国而进攻苏联？_!_
+6552354262494478596_!_110_!_news_military_!_欧盟国家联合研制台风战斗机，为何法国要独立使用阵风战机？两款战机谁更强一些？_!_
+6552347705929105668_!_110_!_news_military_!_日本不害怕东京大轰炸，为何两颗原子弹就投降了？_!_
+6552417421536264455_!_110_!_news_military_!_为什么打仗时，千万不能打跳伞的飞行员？_!_
+6552284434077319687_!_110_!_news_military_!_特大惊喜！中国直18率先登上002航母，美国大呼：剩下时间不多了_!_中国海军,002型航母,海军,系泊试验,直-18
+6552340779816190468_!_110_!_news_military_!_世界十大最强运载火箭排行榜 中国四箭上榜_!_质子M,中国有用载荷,载荷,逃逸塔,联盟号
+6549481638391185671_!_110_!_news_military_!_仅凭苏联一个国家，真的可以打败纳粹德国吗？_!_
+6552307714767192584_!_110_!_news_military_!_美军一老战斗机，仍完胜苏-27！要彻底改进将脱胎换骨_!_AESA,F-15J,苏-27,F-22,战斗机,AMRAAM,导弹,F-15,F-35
+6552379667343474957_!_110_!_news_military_!_为什么军人打电话时不能说方言？_!_
+6552483518146937092_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552035482195198467_!_110_!_news_military_!_俄罗斯后院的战争氛围又变得日渐浓烈，俄英两国这一动作太危险！_!_黑海,俄罗斯,侦察机,乌克兰,北约
+6552409124942905613_!_110_!_news_military_!_如果马其顿军队遭遇大秦军队，谁会赢得胜利？_!_
+6552276994808087053_!_110_!_news_military_!_高新6号再出猛照！_!_高新6号,P-3C,反潜巡逻机,探测仪,运-8Q反潜机
+6552142281112027396_!_110_!_news_military_!_联合国安理会改革废除一票否决权，会导致什么后果？_!_
+6552277847333929486_!_110_!_news_military_!_全球最顶尖的 无人机_!_星球大战,RQ-4A,无人机,全球鹰,以色列,捕食者,攻击机
+6552275955614744846_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552393755435991303_!_110_!_news_military_!_地雷为什么不是一踩就爆，而是还要松开腿才会爆呢？_!_
+6552380519311475214_!_110_!_news_military_!_半自动步枪与全自动步枪的区别 居然还有人不知道！_!_AK-47,M16,自动步枪,加兰德,发射药,子弹,子弹上膛
+6552445591379509773_!_112_!_news_travel_!_祖国的绿水青山处处是美景_!_祖国,泸沽湖,刘文彩,乌镇,黑龙江
+6552468677290099204_!_112_!_news_travel_!_打开武汉全域旅游的“金钥匙”来了_!_服务指南,循礼门站,指南,武汉,大武汉
+6552414180975051271_!_112_!_news_travel_!_中国这个地方居住有5万韩国人，都快被韩国人承包了 就在北京_!_韩国,被韩国,韩国有,当地人,韩国人
+6551720914994594307_!_112_!_news_travel_!_遂溪孔圣山——阿尼亚房车伴我走天涯_!_孔圣山,曲阜,孔圣庙,孔庙,孔子
+6552391215617147143_!_112_!_news_travel_!_去日本的邮轮游需要5万的资产证明吗？_!_
+6552392977392599309_!_112_!_news_travel_!_山东乳山是联合国命名的最适合人类居住的地方，为什么不强化生存条件？_!_
+6552457185752776973_!_112_!_news_travel_!_为什么许多中国人喜好去泰国，而不去三亚？_!_
+6552348837799789060_!_112_!_news_travel_!_山东彻底火了！济南五一纳客122.72万人，泰山一天接待8万人……_!_济南,趵突泉,芙蓉街,千佛山,泰山景区
+6552414729237692680_!_112_!_news_travel_!_佳能70-300的红圈头值得购买吗？旅游拍风景如何？_!_
+6552429443745317123_!_112_!_news_travel_!_上饶有什么好玩的地方？_!_
+6552378314558800392_!_112_!_news_travel_!_家乡山河之《春意盎然》！_!_张广才岭,长白山,吉林市,五常市,哈尔滨
+6552351828829274627_!_112_!_news_travel_!_老挝一座袖珍的老皇城，在西方游客眼中是“世外桃源”_!_世外桃源,老挝,琅勃拉邦,小山城,农夫车
+6552334141403169284_!_112_!_news_travel_!_历时三年，昆明文庙即将开门迎客！_!_云南,棂星门,昆明文庙,大成殿,大成门
+6552371533296173316_!_112_!_news_travel_!_成都有哪些值得推荐的景点或旅游线路？_!_
+6552300408486756877_!_112_!_news_travel_!_外媒称海南将成中国对外开放新窗口：繁荣程度或超越香港_!_自由贸易港,海南经济特区,海南,环岛高铁,海南省
+6552448365739639054_!_112_!_news_travel_!_四川游客张某带着六名女子在三亚沙滩烧烤被撒沙，这事你怎么看？_!_
+6552310274030829837_!_112_!_news_travel_!_日本茨城机场附近有什么能玩的地方么？_!_
+6552461004679152142_!_112_!_news_travel_!_曾经的江南十大名镇之首：如今门票0元，却依然少有游客知道这里_!_古镇,大名镇,特色美食,浙江,旅行者
+6551649228337185287_!_112_!_news_travel_!_震撼：30张世界级照片，你可能连一张也没见过_!_尼拉贡戈火山,火山喷发,熔岩湖,大自然,刚果民主共和国
+6552298979214754051_!_113_!_news_world_!_如果特朗普遭弹劾下台，谁会来接替他当总统？_!_
+6552345320557117959_!_113_!_news_world_!_到底谁才是霸权？外交部这个回应非常霸气，直接硬怼西方两大国！_!_霸权,霸权主义,马克龙,外交部,华春莹
+6552427718540001543_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552461295914844420_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552447694101545224_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552128490135617799_!_113_!_news_world_!_去日本旅游，有哪些值得推荐的景点？_!_
+6552463175348912647_!_113_!_news_world_!_此淡水湖很小，但却是此强国的生命源泉，为了它，不惜与世界为敌_!_中东战争,叙利亚,以色列,戈兰高地,加利利海
+6552345116864938500_!_113_!_news_world_!_政府资助 社会支持——挪威全民参与“海滩清洁日”活动_!_海滩清洁日,奥斯陆,海德,海洋垃圾,挪威
+6552262088318779656_!_113_!_news_world_!_为什么美国人的胡子那么浓密？_!_
+6552325671111098894_!_113_!_news_world_!_《羊脂球》：比鬼神更可怕的，是人心_!_法国,羊脂球
+6552442663507853572_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552309284657431054_!_113_!_news_world_!_韩青瓦台：韩美领导人将于22日举行会谈_!_文在寅,韩国总统,特朗普,青瓦台,韩美
+6552275208097497357_!_113_!_news_world_!_如果证实反政府武装使用化学武器，美英法会被追责吗？_!_
+6552482051944087822_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552124284058206723_!_113_!_news_world_!_2018年诺贝尔文学奖被取消！关于诺贝尔奖的六件有趣事实_!_物理学奖,诺贝尔文学奖,诺贝尔奖,诺贝尔和平奖,拉尔夫·斯坦曼
+6552360152446534147_!_113_!_news_world_!_怀孕6周不准堕胎 爱州出台最严格堕胎禁令_!_禁止堕胎,独立宣言,密西西比州,阿肯色州,大法官,州长
+6552320185288622344_!_113_!_news_world_!_如果美国打伊朗，最快可以多久解决战斗？_!_
+6552400253268525315_!_113_!_news_world_!_孕妇可不可以擦隔离霜呢？_!_
+6552366212204986632_!_113_!_news_world_!_2018年美国中期选举，你认为特朗普会下台吗？_!_
+6552427926090940941_!_115_!_news_agriculture_!_河蟹池塘水草疯长，上浮及烂草等现象防控方法_!_生物制剂,河蟹,光合作用,伊乐藻,水草
+6552361173403042308_!_115_!_news_agriculture_!_坐果率低，树势衰弱，绿盲蝽 当前大樱桃管理技术要点_!_绿盲蝽,生长点,大樱桃,坐果率,授粉树
+6552291854434435591_!_115_!_news_agriculture_!_2018农村有“三块地”的农民要富起来了，一定要守住！_!_承包地,宅基地,农村,三权分置,经营权
+6552488739493380360_!_115_!_news_agriculture_!_假如回到农村生活，你该怎么养活自己？_!_
+6552305247144903171_!_115_!_news_agriculture_!_钱的妙用——有钱能使鬼推磨】接上_!_李生
+6552241251461103886_!_115_!_news_agriculture_!_玉米种子什么牌子的好？_!_
+6552465381808669187_!_115_!_news_agriculture_!_故乡岚县，四月的农家小院早晨_!_农家小院,我的故乡,农家,岚县,小院
+6552446888891646471_!_115_!_news_agriculture_!_互联网+农业 村网通有哪些优势？跟其它类似平台有什么区别？_!_电商网,村站头条,村网通,各地农村门户网站,地区运营负责人,农村,农村政务
+6552278788179231245_!_115_!_news_agriculture_!_大数据+农业 山间“野果”走出“逆袭路”_!_猕猴桃,大数据,修文县,香巴湖村,周珍品
+6552477028711399939_!_115_!_news_agriculture_!_鸡群里的三代同堂——腺肌胃炎_!_胃炎,免疫抑制,腺肌胃炎,鸡内金,三代同堂
+6552366659158409485_!_115_!_news_agriculture_!_种了一亩地蒲公英，该怎么卖？_!_
+6552292468585398541_!_115_!_news_agriculture_!_想开水果店，如何找货源比较好？_!_
+6552286685420339719_!_115_!_news_agriculture_!_2018年深松整地补贴最高每亩60元，哪些地方可以领取补贴？_!_松整地作业补贴,农机,深松整地作业补贴,农机深松整地作业,作业补贴
+6552392307964903949_!_115_!_news_agriculture_!_贝店宣布“一县一品”助农计划，一年将打造100个特色产品产业带_!_助农,贝店,电商平台,县一品,社交电商
+6552299849893544462_!_115_!_news_agriculture_!_茄果类高产第二步_!_水溶肥,茄果类,保花保果,花芽分化
+6552253787640168968_!_115_!_news_agriculture_!_杜清洁：从木工到致富领头羊_!_仙人掌,马甲镇,仙人球,发财树,林子良
+6552485334091825667_!_115_!_news_agriculture_!_农村一种“无心菜”，生长环境水陆通吃，北方吃梗，南方吃叶_!_吃无心菜,蔬菜,北方人,蕹菜,南方人
+6552762979580903950_!_115_!_news_agriculture_!_中国每年进口1.3亿吨粮食 可你知道这些粮食怎么消耗掉的吗_!_粮食加工,我国粮食
+6552264043812356611_!_115_!_news_agriculture_!_乡村教师如何增加自己的收入 看网友们都是如何做的_!_泥水匠,农村,教师,乡村教师
+6552447930605764871_!_116_!_news_game_!_王者荣耀一些知名主播为什么不选择加入职业战队？_!_
+6552399430228640263_!_116_!_news_game_!_Uzi亿元续约虎牙直播，明晚虎牙开播倒计时，各种好礼等你来拿_!_MSI,游戏,RNG,直播间,UZI
+6552250401813430542_!_116_!_news_game_!_怎么看微博大V爆料王者掉帧卡顿等问题是因为手机品牌保护费交的不够？_!_
+6552448819076792846_!_116_!_news_game_!_真·国服第一诺手问世？余小C韩服王者600点诺手豪取五杀！_!_国服第一诺手,诺手,韩服王者600,韩服
+6552444241195303431_!_116_!_news_game_!_RNG官博放出最新片花，唯独少了正在直播的Uzi_!_MLXG,RNG,Uzi,MSI,战队
+6552433592771805699_!_116_!_news_game_!_LOL：亚索第一次在2018MSI登场？KBM完美应对击败DW！_!_KBM,DW,奥拉夫,妖姬,亚索
+6552371075286565390_!_116_!_news_game_!_小智三年坐稳全民一哥，而她成为一姐只用了三个月！_!_你是谁,全民直播,游戏,小智,英雄联盟
+6552315003397997059_!_116_!_news_game_!_Lol里支援速度最快的3个英雄，死歌和慎上榜，第1个对面看到就烦_!_Lol,英雄联盟,SKT,阿卡丽,麻辣香锅
+6552374604055183879_!_116_!_news_game_!_王者荣耀：这四名英雄必被ban！荣耀王者到底有多讨厌他们？_!_Ban,英雄上场,ban掉,Pick,宫本武藏
+6552130348258427395_!_116_!_news_game_!_DNF“玩家花钱在网上买了个号”为了防止找回，直接毁号！_!_DNF,游戏,玩家
+6552253144632393997_!_116_!_news_game_!_英雄联盟闪电狼FW战队到达柏林准备MSI，他们进入正赛的可能性多大？_!_
+6552354859062919688_!_116_!_news_game_!_王者荣耀，哪吒唱起五环之歌，安琪拉摘星星！_!_安琪拉,哪吒
+6552463388360835591_!_116_!_news_game_!_王者荣耀：小学生到底该不该出现在王者峡谷？看完这四点你就明白_!_王者荣耀,王者峡谷,送人头,小学生,LOL
+6552360206888599821_!_116_!_news_game_!_剑仙和张大仙有什么不同呢？_!_
+6552392001952678158_!_116_!_news_game_!_预算7000元左右，是买游戏本还是主机？_!_
+6552397936179806467_!_116_!_news_game_!_长期打游戏对你有什么影响？_!_
+6552335104717357319_!_116_!_news_game_!_300英雄狗年会有活动吗？_!_
+6552382760361656835_!_116_!_news_game_!_dnf童话妖精养成记活动怎么玩？妖精成长积分可兑换奖励介绍_!_童话,dnf,小妖精
+6552387140267803149_!_116_!_news_game_!_王者荣耀：S11赛季特别要注意的事情，敢说这4个字，直接扣分封号_!_王者荣耀,刷国服
+6552404358628114695_!_101_!_news_culture_!_朱标如果不死，会是朱棣的对手吗？_!_
+6552159002522812679_!_101_!_news_culture_!_上联：江中鱼东游西藏，下联怎么对？_!_
+6552351630522581256_!_101_!_news_culture_!_上联：风摇翠柳丝丝舞，怎么对下联？_!_
+6552393015829201155_!_101_!_news_culture_!_上联：鸡鸣犬吠山村美，如何对下联？_!_
+6552264878592098573_!_101_!_news_culture_!_白纸黑字写赤橙黄绿青蓝紫。怎么对下联？_!_
+6552231947018436868_!_101_!_news_culture_!_老子更伟大还是孔子更伟大？_!_
+6552362623852085518_!_101_!_news_culture_!_《红楼梦》中，“金陵十二钗”的命运能否被改变？_!_
+6552299969972273668_!_101_!_news_culture_!_手工串珠装置作品_!_艺术家,Lou,视觉艺术家,Liza
+6552294190531739917_!_101_!_news_culture_!_水浒传挺好，方腊传怎么样？_!_
+6552170931018531080_!_101_!_news_culture_!_顾城算不算是一个诗人？_!_
+6552463468639813896_!_101_!_news_culture_!_下联：万户山城飘炊烟，如何对上联？_!_
+6552259628304957704_!_101_!_news_culture_!_青龙戏白虎，江涛乐夜风。如何接句？_!_
+6552282936131977485_!_101_!_news_culture_!_上联:落笔驚风雨，怎么对下联？_!_
+6552312782614692366_!_101_!_news_culture_!_毛笔字写不好！快看看这位美女的字咋写的，字比人都美_!_狼毫,书法,毛笔字,善琏湖笔,毛笔
+6552489556740931848_!_101_!_news_culture_!_上联：夜阑月下溪柳静，下联怎么对？_!_
+6552247268513153288_!_101_!_news_culture_!_上联：半壶浊酒邀谁饮，如何对下联？_!_
+6552314195646349571_!_101_!_news_culture_!_上联：谁在天上布风雨，胆大包天，如何对下联？_!_
+6552419919969911309_!_101_!_news_culture_!_美到令人窒息的纪录片，央视7集人文纪录片《极地》，推荐给大家_!_杨烁,曾海若,纪录片,民族文化,天彩,极地
+6552469367475405319_!_101_!_news_culture_!_87版陈晓旭真实的生活照，平平淡淡才是真，不语不笑能留人_!_林黛玉,陈晓旭
+6552269253855150340_!_102_!_news_entertainment_!_合伙人排挤你会有哪些表现？_!_
+6552289703469515271_!_102_!_news_entertainment_!_网上爆红的昆池岩，到底有多可怕？_!_头号玩家,恐怖片,信号,毛骨悚然,韩国,昆池岩精神病院
+6552462776520933891_!_102_!_news_entertainment_!_王菲真人秀片酬近5亿？製作单位：传闻不是事实_!_综艺节目,片酬,陈家瑛,湖南卫视,王菲
+6552836848484352526_!_102_!_news_entertainment_!_baby张靓颖关注蔡徐坤，小奶狗人缘圈子越来越大_!_TFBOYS,蔡徐坤,两首歌,张靓颖
+6552484329732178189_!_102_!_news_entertainment_!_如何以《镜花水月》写七绝藏头诗？_!_
+6552365149800366344_!_102_!_news_entertainment_!_《雍正王朝》和《康熙王朝》，你更喜欢哪部电视剧？_!_
+6552451293305111047_!_102_!_news_entertainment_!_「奶头乐理论」｜我们将毁于我们热爱的东西_!_奶头乐理论,边缘人,步步惊心,点击率,花边新闻
+6552284007801815559_!_102_!_news_entertainment_!_58岁毕福剑低调现身，没毕姥爷的《星光大道》还习惯吗？_!_毕姥爷,星光大道,毕福剑
+6552320583206437127_!_102_!_news_entertainment_!_《请回答1988》传递给大家的三观，真的是充满正能量的吗？_!_
+6552411352072192519_!_102_!_news_entertainment_!_56岁岳秀清和56岁吴刚近照，两人恩爱了28年，21岁儿子长成这样_!_铁人,漂亮妈妈,阮玲玉,天下第一楼,茶馆,张大民,岳秀清,大器晚成
+6552469859618259459_!_102_!_news_entertainment_!_秦海璐出席活动，网友：个性太强了胸口有只小手铐？_!_秦海璐出席活动秦海璐出席,秦海璐
+6552489354114105860_!_102_!_news_entertainment_!_王志文的初恋，冯小刚都曾向她表白，今49岁不婚做自己的公主！_!_花少,王志文,老炮儿,花儿与少年,冯小刚,陈凯歌,24小时·末路重生,妹妹们,许晴
+6552389280528859406_!_102_!_news_entertainment_!_你心目中的华语最佳男女歌手是谁？_!_
+6552156673249640963_!_102_!_news_entertainment_!_两个女神明星产后归来，一个万丈光芒，另一个却饱受网友吐槽_!_女明星,湖南卫视,孤芳不自赏,奔跑吧！兄弟,万丈光芒,快乐大本营,谢娜
+6552370981401264653_!_102_!_news_entertainment_!_刘涛孙俪这些大牌亿万身价，都比不上她？这个女星终于熬成一线了_!_女星,甄嬛传,刘涛,赵丽颖,女明星,孙俪
+6552448971569103374_!_102_!_news_entertainment_!_聂风、步惊云、秦霜都有招牌武功，那跟他们一起长大的断浪呢_!_排云掌,步惊云,天霜拳,断浪,雄霸,风云
+6552297063508345351_!_102_!_news_entertainment_!_谢娜孕后含泪首录《快本》快乐家族合体惹人泪_!_何炅,快乐大本营,快乐家族,谢娜
+6552325378877161742_!_102_!_news_entertainment_!_李小龙创立的截拳道为何没有在大众普及？_!_
+6552390996020167171_!_102_!_news_entertainment_!_盘点那些带上眼镜的男明星，感觉鹿晗颜值瞬间提升一个档次_!_
+6552362630785270280_!_103_!_news_sports_!_皇马会搁置欧冠，全力狙击巴萨的不败夺冠吗？齐达内公开表态_!_巴萨,欧冠,齐达内,皇马,国家德比
+6552331146623975943_!_103_!_news_sports_!_恩比德：我们掌控比赛，只是在最后犯了错_!_恩比德,下一场
+6552490062171341325_!_103_!_news_sports_!_2017/18赛季英超争四，利物浦vs切尔西最后的决战_!_欧冠决赛,利物浦,英超,皇马,切尔西
+6552461402768933123_!_103_!_news_sports_!_马刺可以拿卡哇伊和雄鹿换字母哥吗？_!_
+6552307834338410756_!_103_!_news_sports_!_为何CBA篮球运动员孙悦的妻子陈露被称为“CBA第一夫人”？_!_
+6552452977318166797_!_103_!_news_sports_!_西班牙国家德比大名单最新出炉，无欲无求之下，皇马巴萨决定死磕，你怎么看？_!_
+6552133490727452941_!_103_!_news_sports_!_郭跃在社交平台上说丁宁，每次输掉比赛都说受伤，对此你怎么看？_!_
+6552469371942338829_!_103_!_news_sports_!_詹姆斯绝杀的最后8秒时间，叫了暂停为什么是后场发球，而不是前场发球？_!_
+6552270132519895300_!_103_!_news_sports_!_骑士目前2比0猛龙，晋级的概率有多大？_!_
+6552445620152435207_!_103_!_news_sports_!_古巴传奇名将路易斯当选FIVB副主席_!_路易斯,古巴女排,古巴,FIVB
+6552314276541890819_!_103_!_news_sports_!_如果刘国梁出国执教你是支持还是反对？_!_
+6552342194546541064_!_103_!_news_sports_!_亿元合同，末节被抛弃，三节仅3分，他该被讽刺还是鼓励？_!_猛龙队,德罗赞,职业生涯,精神疾病,半决赛
+6552397584973955588_!_103_!_news_sports_!_温格主场末战赛前寄语：团队足球是一种艺术_!_佩蒂特,阿森纳,团队足球,温格,伯恩利
+6552286353827037699_!_103_!_news_sports_!_朗多心机man?意图垫脚库里不能忍！_!_库里,科尔,录像带,勇士队,金州勇士
+6552342706469732622_!_103_!_news_sports_!_足球比赛中守门员什么情况下可以用手，什么时候又不能？_!_
+6552451211243553293_!_103_!_news_sports_!_当浓眉哥轰下33分18板时，你才会发现隆多的厉害之处_!_季后赛,nba官网,隆多,NBA,浓眉哥
+6552396270873346308_!_103_!_news_sports_!_李易峰晒照支持詹姆斯，他是詹姆斯的球迷吗？_!_
+6552159353804161294_!_103_!_news_sports_!_如何评价韩宇夺得《这！就是街舞》冠军？_!_
+6552491383796531715_!_103_!_news_sports_!_谁说哈登交女友状态就会下滑，哈登季后赛数据打脸太多人！_!_哈登,季后赛,火箭,夜店
+6552318489820922120_!_103_!_news_sports_!_刘伟解说乒乓球比赛的水平如何？_!_
+6552305797387256071_!_104_!_news_finance_!_用手机怎么赚钱？_!_
+6552269398101459207_!_104_!_news_finance_!_一斤两块五的青菜人都买的吃，为什么两块五的股票没人买呢？_!_
+6550858129825857795_!_104_!_news_finance_!_企业持有巨额现金，是好还是不好？_!_
+6552154060798034179_!_104_!_news_finance_!_潘刚32岁，出任伊利集团总裁是怎么做到的？_!_
+6552477034273046798_!_104_!_news_finance_!_美国工薪家庭基本支出和改善性支出都有哪些？_!_
+6552290362176569864_!_104_!_news_finance_!_先锋股票池（5月7日）_!_股票池
+6552480012665094669_!_104_!_news_finance_!_最高法：上市公司对外担保合同不因违反对内程序规定而无效_!_保证合同,股东会决议,法定代表人,增资扩股协议,合同效力,股东会,贤成矿业,承诺保证函,第三人
+6552294744150508040_!_104_!_news_finance_!_美天然气NaturalGas中短线交易策略及技术走势MT4解盘五月三日_!_MACD,K线,一目均衡表,周期线,图MACD,阻力位
+6552364321131725319_!_104_!_news_finance_!_巴菲特说美中关系可以双赢_!_内布拉斯加州,伯克希尔,沃伦·巴菲特,美中关系,奥马哈市
+6552274974890000909_!_104_!_news_finance_!_4部门发文规范民间借贷：禁暴力催收、校园贷_!_高利贷,非法拘禁,银行业金融机构,民间借贷,校园贷
+6552363863298277902_!_104_!_news_finance_!_「5.6全球资讯」乌克兰出现首个全村拥有数字货币的村庄_!_虚拟货币,数字货币,阿塞拜疆,区块链,比特币
+6552005659087340039_!_104_!_news_finance_!_详解简单精准的抄底技巧，真正搞懂的寥寥无几_!_成交量,抄底,底部,股价,乖离率
+6552359120941351428_!_104_!_news_finance_!_EPFR：中国债券基金吸金量触及11周高位_!_股票,EPFR,epfr,债券基金,QFII
+6552298842899874307_!_104_!_news_finance_!_国产芯片第一龙头势如破竹，神秘巨资增持8900万股，18年能否崛起_!_集成电路,挠性电路,高技术,丹邦科技,柔性封装基板
+6548738415100166404_!_104_!_news_finance_!_买的股票每次卖后面都会涨，一拿着就肯定跌，怎么回事？_!_
+6552374939297513998_!_104_!_news_finance_!_巴菲特与科技公司的爱恨情仇：对贝索斯和马斯克的态度截然相反_!_太阳能,亚马逊,BNSF,伯克希尔,谷歌,公司债,棱镜,巴菲特,伯克希尔哈撒韦,摩根大通,Geico,特斯拉,马斯克,贝索斯
+6552249063478132996_!_106_!_news_house_!_一般上海家庭资产都是过千万吗？_!_
+6552346174701961485_!_106_!_news_house_!_现在小县城房子6000多，适合买房吗？_!_
+6552446188405129480_!_106_!_news_house_!_湘潭的房价趋势如何？_!_
+6552269008337371395_!_106_!_news_house_!_任志强说现在是“抄底时机”，购房者该不该信？_!_
+6552372006824706311_!_106_!_news_house_!_贷款利率上涨下的房价上涨，真的刚需吗？_!_
+6552427773233725703_!_106_!_news_house_!_造成目前房价过高的推手到底谁？_!_
+6552266224141598980_!_106_!_news_house_!_长春东湾半岛房子怎么样？_!_
+6552417232347988231_!_106_!_news_house_!_成都有富人区吗？在哪里？_!_
+6552328003857154563_!_106_!_news_house_!_国际酒店公寓楼上就是你的家_!_买房,公寓楼,LOFT,国际酒店
+6552292929640071438_!_106_!_news_house_!_建房屋第一层最高可以多少米？_!_
+6552327398161908231_!_106_!_news_house_!_福州市中心、奥体两新盘备案名出炉！鼓楼新项目售价仅17800元/平_!_厦门住宅,奥体,华润,福州市,福州
+6552368012446401038_!_106_!_news_house_!_淄博人速看！不管淄博人买不买房，未来都会出现这两件事！_!_房价,房产,楼市,买房,楼盘
+6552352547602956813_!_106_!_news_house_!_不经过夫妻同意，卖房，危害太大了？_!_另一方,卖房,委托书,购房,违约责任
+6552315160478875908_!_106_!_news_house_!_“六个钱包”买房为什么会被误读？_!_
+6552436067557966083_!_106_!_news_house_!_苏通园区跟锡通园区到底哪个值得买房？_!_
+6552382630610862343_!_106_!_news_house_!_买房子交了首付后交不起房贷怎么办？_!_
+6552378660660183303_!_106_!_news_house_!_未来房价会跌到什么程度？_!_
+6552340613180686856_!_107_!_news_car_!_哎呦哎～聪明老公给车里换上这样的冰丝坐垫，这个夏天凉爽又舒适_!_四季通用,广汽本田,冰丝,冰丝汽车坐垫,SUV
+6552336562875531790_!_107_!_news_car_!_20万元落地长城VV7S旗舰型，开了一万公里后车主自信的晒出了油耗_!_WEY,绅士化,国产车,长城,SUV
+6552273386062479879_!_107_!_news_car_!_10万元级别！最值得看的五款SUV来了，内饰上采取了细节上的优化_!_SUV,北京车展,T700
+6552292879446835715_!_107_!_news_car_!_你知道卖不掉的库存车，4S店是怎么处置的吗？_!_雨刮器,经销商,4s店,库存车,汽车历史
+6552435025558634759_!_107_!_news_car_!_佛山一小车没给奔驰车逆行让路，竟遭奔驰车主拍引擎盖辱骂，对此你怎么看？_!_
+6551920890810991111_!_107_!_news_car_!_隧道行车你做对了吗？通行技能了解一下_!_隧道行车,报警灯,隧道,安全地带,行车
+6552449238666576387_!_107_!_news_car_!_奥迪Q7是一款强调舒适性的SUV，高科技和豪华品质巧妙地融为一体_!_奥迪Q7,4s店,Q7,奥迪,奥迪q7,SUV,奥迪公司
+6552288436768735757_!_107_!_news_car_!_教练讨要红包真是太气愤了，还能不能好好学车了？_!_教练收红包,红包,驾照考试
+6552310042698187267_!_107_!_news_car_!_某些合资汽车品牌为何陆陆续续退出中国市场？_!_广汽菲亚特,比亚迪,菲亚特,合资车,克莱斯勒,长安铃木,国产车
+6552313951261032967_!_107_!_news_car_!_“敢性人生，无惧路上”——哈弗将开启粉丝品牌新时代_!_哈弗H6,特斯拉,哈弗,Coupe,哈弗H9,哈弗H2,哈弗H7,HAVAL,哈弗H8,哈弗粉丝,马自达,哈弗H1
+6552336167541408013_!_107_!_news_car_!_能用电磁力驱动汽车吗？_!_
+6552293329600512516_!_107_!_news_car_!_自动驾驶又出车祸，这项新技术到底靠不靠谱？！_!_Chandler,Uber,无人驾驶,克莱斯勒,Pacifica,Waymo自动驾驶汽车
+6552495888164454920_!_107_!_news_car_!_最厚道丰田又降价！曾卖15万比朗逸风光，现配CVT6万起比捷达便宜_!_威驰,起亚K2,卡罗拉,捷达,朗逸,丰田,CVT6,瑞纳
+6552275033144688909_!_107_!_news_car_!_汽车在正常情况下排气管发黑，是油气配比问题吗，具体动力与燃油经济性协调关系是怎样的？_!_
+6552257267033440520_!_107_!_news_car_!_长城哈弗H6有什么缺陷吗？_!_
+6552335186938298628_!_107_!_news_car_!_求问，25万左右落地suv，性能不要求太高，但要求舒适型高，哪款比较合适？_!_
+6552285584134504973_!_107_!_news_car_!_男人一旦到了50！就该早点备上这样个性狠物，超好玩_!_手机支架,防撞条,手电钻,剃须刀,疏通器
+6552286492415230467_!_107_!_news_car_!_钛金灰瑞虎8，传承家族最新“虎踞”式设计风格，凸显威武霸气_!_SUV,速双离合,瑞虎,瑞虎8
+6552238103770169869_!_108_!_news_edu_!_吉林工程职业学院携手凤凰数字媒体教育全力推进产学合作协同育人_!_凤凰数字媒体教育,媒体教育,职业学院党委副书记,吉林工程职业学院,工程职业学院党委副
+6552278236657615117_!_108_!_news_edu_!_2018年高考陕西文科550能上哪些师范院校？_!_
+6552459085898318343_!_108_!_news_edu_!_河南师范大学新联学院2018届美术学毕业生画展部分作品展示_!_河南师范大学新联学院,美术学,画展河南师范大学新联学院2018,美术学毕业生画展河南师范大学新联,画展
+6552389384551793156_!_108_!_news_edu_!_我的生活我做主_!_比赛现场剪影,特殊教育学校,系鞋带,王卫平,生活技能
+6552299352092574216_!_108_!_news_edu_!_荥经县胡长保幼儿园开展家长开放日活动_!_胡长保,荥经县,幼儿园,家长开放日
+6552496686340506116_!_108_!_news_edu_!_高考即将到来，高考对考生而言有多麽的重要，你怎么看呢？_!_教育部考试中心
+6552391924215448068_!_108_!_news_edu_!_女人的性格决定了她的婚姻幸福程度_!_麻辣烫,小锦,性格,学不会,文俊
+6552343758334066958_!_108_!_news_edu_!_2018年四川理科考生500分左右能报哪些学校？_!_
+6552415307556716807_!_108_!_news_edu_!_学生托管班机构怎样加强师资力量的培训？_!_
+6552364037571608846_!_108_!_news_edu_!_为什么要有中考？_!_
+6552352639529517320_!_108_!_news_edu_!_对于不爱英语的孩子怎么让孩子学习英语？有靠谱点的少儿英语培训机构推荐吗？_!_
+6552072043670536718_!_108_!_news_edu_!_高中数学“曲线的对称性”知识要点归纳和解题方法总结！_!_高中数学
+6552261988037165575_!_108_!_news_edu_!_忧虑质疑才能迈向未来：哈佛大学350周年校长讲话稿通篇证明_!_林校长,洛厄尔,高等教育,生物工艺学,讲话稿,学术,极限点,哈佛大学,哈佛,毕业生
+6552456419902226695_!_108_!_news_edu_!_南昌大学怎么样？_!_
+6552311209197044232_!_108_!_news_edu_!_对台湾政治纷争失望！“教育部”拔管 教师爆“逃难潮”_!_台湾,台大,拔管,教育部,逃难潮
+6552270329215975950_!_108_!_news_edu_!_国外部 留学家长口语俱乐部毕业典礼_!_粑粑麻麻,毕业典礼,新东方
+6551938980844143108_!_108_!_news_edu_!_高清直播｜佛山一中校长解读最新招生政策，支招志愿填报_!_名师堂,校长
+6552251767835329037_!_108_!_news_edu_!_“鸿鹄之志”北大校长读错了不重要，这个成语的典故才值得反思！_!_林建华,吴广,鸿鹄之志,农民起义,陈胜
+6552274842748453380_!_108_!_news_edu_!_武汉理工大学2018年录取分数线 历年录取最高最低分_!_大数据,指导中心大数据来源,武汉汽车工业大学,武汉理工大学,广西高考志愿
+6552729100463636995_!_109_!_news_tech_!_全球智能手机 2018 Q1 出货排名出炉，苹果包揽前四_!_Android,iPhone 8,出货量,智能手机,iPhone X
+6552419429060182285_!_109_!_news_tech_!_海尔专营店和海尔专卖店有什么区别？_!_
+6552305736402075908_!_109_!_news_tech_!_移动宽带为什么那么贵？_!_
+6552327362850062862_!_109_!_news_tech_!_机器人送货渐唱主调 电商巨头推进末端配送无人化_!_机器人,无人配送车,无人化,无人机,末端配送
+6552452241431724301_!_109_!_news_tech_!_路由器长时间不关，会造成网络信号不好吗，为什么？_!_
+6552487698509070862_!_109_!_news_tech_!_2018年最新靠谱网赚项目！日赚300元_!_天猫,优惠券,互联网,手机app,马云
+6552291532488048900_!_109_!_news_tech_!_中国人一年不买苹果手机，苹果会破产吗？_!_
+6552307555538829828_!_109_!_news_tech_!_区块链+新零售：是新经济风口，或是简单联姻，网友：新时代必然_!_小米手机,零售,阿里巴巴,新经济,区块链加新零售
+6552298327004676616_!_109_!_news_tech_!_腾讯被批没有梦想，马化腾破天荒火速回应：我的理想并不是想赚钱_!_腾讯没有梦想,马化腾,腾讯
+6552427058507546893_!_109_!_news_tech_!_假如有个好产品，如何找一个好的营销团队？_!_
+6552468611275948295_!_109_!_news_tech_!_哪个共享汽车app收费比较便宜？_!_
+6552286476221022733_!_109_!_news_tech_!_如果有时光机 你最想去和谁见面？马云想见的是他们_!_爱因斯坦,特拉维夫,以色列,阿里巴巴,马云
+6552371377897210126_!_109_!_news_tech_!_从20%暴跌到0.8%仅用五年，三星手机为何会在中国遇雪崩式溃败？_!_
+6552266886342509063_!_109_!_news_tech_!_腾讯没有梦想_!_微视,马化腾,刘炽平,王者荣耀,飞车炫舞,任宇昕,腾讯,恋与制作人,光荣使命,张小龙,旅行青蛙,腾讯微博,腾讯传,SNG
+6552248011827708174_!_109_!_news_tech_!_小米累计“亏损”如此严重，为什么还能提交港股上市？_!_
+6552376495929557517_!_109_!_news_tech_!_贵阳大数据、人工智能等前沿技术发展，先行优势已经突显_!_人工智能,大数据,前沿技术,创业邦,贵阳
+6552095491012690183_!_109_!_news_tech_!_小米要上市了，有十万块钱怎么能买到小米的股票？_!_
+6552391453136388611_!_109_!_news_tech_!_小米将在英国及欧洲开售：美国或是下一个目标市场_!_李嘉诚,小米,路透社,雷军,小米手机
+6552450784640893188_!_109_!_news_tech_!_怎么评价潘乱写的《腾讯没有梦想》？_!_
+6552456145733157134_!_110_!_news_military_!_获得过美国荣誉勋章的美国大兵为什么不能重新回到战场？_!_
+6552386864148381966_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552491755793547524_!_110_!_news_military_!_法国和英国联合研制的协和式超音速客机为什么退出运营？_!_
+6552406339417538819_!_110_!_news_military_!_德国现役台风战机只有4架具备实战能力，这说明了什么？_!_
+6552489472338952461_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552431836688023821_!_110_!_news_military_!_你怎么看待以色列总理出席俄罗斯5月9日的阅兵并会晤普京？_!_
+6552470708235010318_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552330475560501518_!_110_!_news_military_!_拿破仑麾下的“老近卫军”有何来历，为何对拿破仑忠心耿耿？_!_
+6552288156228518407_!_110_!_news_military_!_中国攻克核心技术！055大驱将用上这款神器，美国感叹：赶不上了_!_军舰,推进系统,中国海军,电磁炮,MK-41,电推进
+6552391052852986119_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552310100910932487_!_110_!_news_military_!_中国灵巧炸弹漫长升级之路-40年努力成世界一流_!_导弹,宝石路,卫星,JDAM,激光制导炸弹
+6552262096761913614_!_110_!_news_military_!_子弹出膛瞬间，手腕急速抖动，枪口恰好形成作用于子弹尾部的力量会发生什么变化？_!_
+6552264011558158599_!_110_!_news_military_!_女乒今天排兵布阵不合理，丁宁昨天刚打硬仗，今天应该打第一单打，你认为呢？_!_
+6551871108226744846_!_110_!_news_military_!_澳洲专家告诉你中国新型导弹为什么能主宰远海吓退他国海空军！_!_S-300PMU,地对空导弹,YJ-12,S-400,航母,SM-2,海空军,歼-11B,海军,歼-15
+6552429631956320782_!_110_!_news_military_!_全军实行体重强制达标制：关乎战斗力_!_共同条令,体重达标,战斗力,阿富汗,体重超标,军人
+6551907472829841933_!_110_!_news_military_!_《红海行动》中的她，没想到竟是释小龙的师妹_!_释小龙,北京体育大学,蛟龙突击队,红海行动,蒋璐霞,战狼,功夫梦
+6552226431244435715_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552253295438594567_!_110_!_news_military_!_“科摩多-2018”多国联演开幕 中国海军编队参加国际海上阅舰式_!_龙目岛,国海军,曾行贱,阅舰式,中国海军
+6552415604962230788_!_112_!_news_travel_!_安徽徐集镇以绿色生态为理念 建成紫荆花特色小镇带动乡村发展_!_紫荆花特色小镇,安徽省,无人机拍摄,六安市,徐集镇
+6552443827586597380_!_112_!_news_travel_!_拍梯田何必去云南，利川处处媲美它……_!_层层叠叠,大自然,吊脚楼,土家族,世外桃源
+6552233389187924494_!_112_!_news_travel_!_雪乡印象，虽然屡屡出现宰客丑闻，风景确实很美_!_洁白如玉,名声大噪,雪乡印象,东方IC,爸爸去哪儿,银装素裹
+6552383561926705412_!_112_!_news_travel_!_国家为什么不允许用红砖建房？_!_
+6552138717279027716_!_112_!_news_travel_!_深圳市宝安区的“海景公园”——西湾红树林公园_!_小梅沙,红树林公园,广深高速公路,黄埔区,西湾
+6552484375408149006_!_112_!_news_travel_!_拉萨义工旅行｜拉萨，从记忆客栈开始！_!_高原反应,拉萨,繁星点点,布达拉宫,纳木错
+6552496451191046660_!_112_!_news_travel_!_好闺蜜一起游玩三亚——看看蓝天碧海！_!_很好吃,三亚湾,三亚,菠萝蜜,小包子,大东海,清补凉,亚龙湾
+6552128827600929028_!_112_!_news_travel_!_行程是十天，东京进出，中途要去京都和奈良然后回东京，如何买交通卡呢？_!_
+6552250424542364167_!_112_!_news_travel_!_平陆小三亚，坡底粮宿滩_!_粮宿滩,平陆县,三亚,平陆
+6552337648864723459_!_112_!_news_travel_!_APEC怀柔会址最佳观览点建成_!_观览点,APEC,雁栖湖,apec,怀柔
+6552290479650636301_!_112_!_news_travel_!_微信史上最囧挑战12关图文攻略 史上最囧挑战12关找不同的猫答案_!_
+6552410445209141517_!_112_!_news_travel_!_武汉东湖绿道开通以后，你去过吗？_!_
+6552468568523407879_!_112_!_news_travel_!_苏锡常最少去的是无锡，小笼包子江南最甜_!_无锡太湖,南长街,无锡,常州,待客之道
+6552323141350195460_!_112_!_news_travel_!_旅行时和异性拼房是一种怎样的体验？_!_
+6552383478149677581_!_112_!_news_travel_!_干货｜持中国大陆护照在新加坡如何申请入台证_!_签证,许可证,台湾,入台证,大通证
+6552278575091810567_!_112_!_news_travel_!_第二次去京都，附近有没有很值得去的地方？_!_
+6552455593775333645_!_112_!_news_travel_!_广西北海和三亚哪个好玩？费用哪个高？_!_
+6552143721058533635_!_112_!_news_travel_!_为什么有人说桂林是下一个凤凰古城？_!_
+6552261537065599240_!_112_!_news_travel_!_如何从大阪去关西机场？_!_
+6552472190413963779_!_112_!_news_travel_!_重庆最火爆的网红民宿——花生民宿系列《别样》_!_北滨路,重庆,小吃街,榻榻米,原研哉
+6552279941621219854_!_113_!_news_world_!_特朗普彰显“商人”智慧，巧施一计就可让俄罗斯陷入泥潭！_!_耶路撒冷,鬣狗,俄罗斯,盟友,叙利亚
+6552461846287221006_!_113_!_news_world_!_如果越南现在仍用中文作为官方语言，国家会是什么情况？_!_
+6552385449631941127_!_113_!_news_world_!_丈夫醉酒强行发生关系不得逞，暴揍妻子被行政拘留3天_!_家暴
+6552326117057888516_!_113_!_news_world_!_世界上最有影响力的国家是哪个，为什么？_!_
+6552246286039384323_!_113_!_news_world_!_普京在刚上台时说要带领俄罗斯成为强国，就目前的形势来看，俄罗斯是否成为了强国？_!_
+6552444059795849476_!_113_!_news_world_!_美国建国不过两百年，为什么会成为世界强国？_!_
+6552483772854436099_!_113_!_news_world_!_如果把美国换成俄罗斯，世界会怎么样？_!_
+6552311465515155716_!_113_!_news_world_!_为什么说民主党很难赢得2018年美国的中期选举了？_!_
+6552240473258328328_!_113_!_news_world_!_美国建国不过两百年，为什么会成为世界强国？_!_
+6552275340926910979_!_113_!_news_world_!_英推出世界首家太空游机构 太空游18.3万英镑起_!_太空酒店,Bark,英国,太空飞行,太空游
+6552490742860742920_!_113_!_news_world_!_美国人没有任何储蓄吗，70％的人连1000块钱都拿不出来是真的吗？_!_
+6552239014185795843_!_113_!_news_world_!_如果日本国土滑向海沟沉没，要求中国人民收留，作为中国人你会接受日本人吗？_!_
+6552394556631941635_!_113_!_news_world_!_日本交通事故率全球最低，他们是怎么做到的？网友：国民素质第一_!_交通事故率,日本,日本全国,交通事故死亡率,闯红灯
+6552363090598429192_!_113_!_news_world_!_哈萨克斯坦发生4.0级地震_!_哈萨克斯坦,中国地震台网,地震,震源深度
+6552242864435233032_!_113_!_news_world_!_如果证实反政府武装使用化学武器，美英法会被追责吗？_!_
+6552434916301209869_!_113_!_news_world_!_日本和以色列为什么会在贫瘠的土地上建立起高度发达的国家？_!_
+6552344094096490765_!_113_!_news_world_!_印度穷人很多，为什么没有起来武装闹革命？_!_
+6552494704171155976_!_113_!_news_world_!_和尚胆大破戒，公然在电梯和女子撒了一波狗粮，让人看傻了眼_!_电梯,小和尚,电梯门
+6552385710337294852_!_115_!_news_agriculture_!_种瓜得瓜，种豆得豆，种发了芽的土豆竟然是这个！_!_农家,土豆丝,稻草堆,发芽土豆
+6552167335149437198_!_115_!_news_agriculture_!_近几年庄稼病虫害增加，适当休耕既可解决地力枯竭的问题，又可减少病虫害的发生，你怎么看？_!_
+6552414078487233032_!_115_!_news_agriculture_!_这种植物，在路边是绿化，上餐桌是美食，进医院就是中药材！_!_鼠尾草,中药材,农人,回羽状复叶,调味剂
+6552151204015636995_!_115_!_news_agriculture_!_朋友种的果树丰收了，简直太诱惑我了，不花钱就能吃到这么多水果_!_果树,猕猴桃,奇异果,杏子,南方北方
+6552239620573102605_!_115_!_news_agriculture_!_田园综合体咋建？山东省率先发布四项标准_!_朱家林,田园,田园综合体建设,田园社区,沂南县,田园社区建设规范总则
+6552315804904325646_!_115_!_news_agriculture_!_世界那么大，独爱庞各庄小西瓜_!_杨锦华,星期六,西瓜,庞各庄,庞各庄西瓜
+6552266206605213966_!_115_!_news_agriculture_!_怎样能提高春雨小麦收割机的速度？_!_
+6552289419435442440_!_115_!_news_agriculture_!_俄罗斯的农作物和蔬菜真的不使用农药化肥吗？_!_
+6552399398368707086_!_115_!_news_agriculture_!_如果宅基地三权分置，你的收益可能不止是你家的宅基地哦！_!_使用权,宅基地三权分置,农村宅基地,所有权,土地权属
+6552286708405109261_!_115_!_news_agriculture_!_三亚独有“牛奶莲雾”上市 本地消费者“一果难求”_!_三亚,长山村,牛奶莲雾,王晓斌,莲雾
+6515569738322543108_!_115_!_news_agriculture_!_柑橘脱毒容器苗栽培技术_!_柑橘脱毒容器苗,栽植,根系,苗木,西充县
+6552277233363321348_!_115_!_news_agriculture_!_2018/5/6今日全国生猪报价_!_生猪,生猪价格
+6552273640719647240_!_115_!_news_agriculture_!_温州“六千六万”行动助力乡村振兴_!_乡村振兴,六千六万,西坑畲族镇,温州,意见,石斛谷
+6552112057326502404_!_115_!_news_agriculture_!_蒜薹收获季，愁坏了庄稼人，本应喜迎丰收却为了四处求人收蒜薹_!_庄稼人,大蒜,蒜薹,蒜薹价格
+6552344131299967492_!_115_!_news_agriculture_!_从一家15年不倒的水果店，看农业经营的5大启示_!_保鲜纸,农业,水果,源兴,差异化
+6552271334787777032_!_115_!_news_agriculture_!_它长相似榴莲，个头大过榴莲，你知道是什么水果吗？_!_菠萝蜜,榴莲
+6552448524540183043_!_115_!_news_agriculture_!_湖南省常德市桃源县农村三层别墅高端现代创意别墅外观更精美_!_桃源县,创意别墅外观,层别墅高端,精美湖南省常德市桃源县农村,湖南省
+6551736872454848782_!_115_!_news_agriculture_!_大丽花的花朵谢了，用剪掉吗？_!_
+6552291135396512270_!_115_!_news_agriculture_!_现在流行的阳台盆栽瓜果，沾土就活，果子爆盆压断枝，多到吃不完_!_水蜜桃,无花果,南方北方,嫁接苗,火龙果
+6552064081807802893_!_116_!_news_game_!_DNF 总算有个号闪了万世三件套 感觉好受伤啊 俺们运气是有多坑啊_!_dnf,游戏
+6524513908470841603_!_116_!_news_game_!_为什么玩穿越火线，有的人打死人后老是喜欢换子弹？_!_
+6552390700271403528_!_116_!_news_game_!_eStarPro vs WF.D前瞻：背水一战，WF.D脱离保级区之路任重而道远_!_老夫子,程咬金,eStarPro,WF.D,孙膑
+6552146570148577800_!_116_!_news_game_!_联发科P60八核让OPPO R15吃鸡更流畅_!_吃鸡游戏,游戏,联发科P60,OPPO R15,吃鸡
+6552316842193453576_!_116_!_news_game_!_梦幻西游这几天连出三把专用，现在已经很难取舍了_!_初始命中,双加属性,大唐官府专用,初始伤害,100级
+6552332047342371341_!_116_!_news_game_!_LOL：麻辣香锅韩服Rank不合成打野刀？电刀巨魔新套路连胜中_!_MSI,韩服,巨魔,麻辣香锅,Rank
+6552422174299259399_!_116_!_news_game_!_稳住，我们能赢！因为我们拥有这种强大的力量_!_团队精神,王者荣耀,老夫子,I Will Carry You,Carry,东皇太一
+6527253239417536781_!_116_!_news_game_!_腾讯是怎样毁掉《斗战神》这款可能成神的作品的？_!_
+6552358480567599619_!_116_!_news_game_!_LOL你不知道的黑科技打法！_!_奶妈,索拉卡,琴女,伤害,ap加成
+6530169419614650632_!_116_!_news_game_!_游戏拳皇各代的剧情都在讲些什么？_!_
+6552452215494148621_!_116_!_news_game_!_《战神4》迷雾之锚获得方法_!_女武神,战神4,BOSS
+6551661650380849667_!_116_!_news_game_!_王者荣耀：国服最强 庄周单排心得，附廉颇出装铭文_!_队友,仙豆手游,决战九霄,廉颇,对面,剑逆八荒,老夫子
+6552464271551234311_!_116_!_news_game_!_《英雄联盟》中被大家称为世界第一adc的uzi和别人有什么不同呢？_!_
+6552295423631950343_!_116_!_news_game_!_绝地求生 手游 如何吃鸡_!_手游
+6552407094476145159_!_116_!_news_game_!_大神带玩，连红_!_
+6552442469902975492_!_116_!_news_game_!_王者荣耀：剑仙138连胜，骚白326连胜，却远不如他的959连胜_!_剑仙,王者荣耀,138连胜,连胜,路人王
+6552160142102626823_!_116_!_news_game_!_来之不易的胜利！这支战队使出浑身解数终于拿下赛季第二胜！_!_马可波罗,暴君,JC,路西法,YTG
+6552445305692881159_!_116_!_news_game_!_《王者荣耀》S11赛季，强势的单带英雄有哪些？_!_
+6543859566096941325_!_101_!_news_culture_!_鲁迅每月工资350块大洋，换成现在的人民币值多少钱？_!_
+6522957178771341828_!_101_!_news_culture_!_中国古代十句最美丽的谎言，以及十句凄美的回复_!_君临天下
+6531136600456299016_!_101_!_news_culture_!_霸气的齐白石说了：卖画不论交情，请照润格付钱_!_艺术界,北京画院,齐白石
+6538341327615033864_!_101_!_news_culture_!_史上最牛逼的一句唐诗，千百年来不断被“抄袭”_!_集句诗,李贺,忆秦娥,行路难,金铜仙人,唐诗,金铜仙人辞汉歌,减字木兰花,七律,蝶恋花,诗鬼
+6551544719204155652_!_101_!_news_culture_!_“天龙三兄弟”乔峰、虚竹和段誉联手能否击败石破天？_!_
+6550236270470103555_!_101_!_news_culture_!_扁鹊失传医书已经出土，历史教科书仍坚持将他删除，真的好吗？_!_敝昔医论,脉数,汉书•古今人表,列子•汤问,蔡桓公,扁鹊,秦越人,扁鹊外经,赵简子,扁鹊内经,史记•扁鹊传,史记志疑,诸病症侯,中医学,难经
+6542259933264478727_!_101_!_news_culture_!_山东小学三年级女孩书法，可比肩当代名家！_!_名家,书法,山东
+6547128557817037326_!_101_!_news_culture_!_《红楼梦》书里的十大百年谜题，红学界竟然无人能解释清楚_!_曹雪芹,秦可卿,宁国府,贾兰,红楼梦,史太君
+6466758384300327182_!_101_!_news_culture_!_四灵到底是麒麟、龙、龟、凤凰还是青龙、白虎、玄武、朱雀？_!_麟凤龟龙,金翅大鹏雕,西游记,麒麟,如来佛祖
+6529662375778320900_!_101_!_news_culture_!_观音本有能力秒杀红孩儿，她为何还要向李天王借天罡刀？_!_孙悟空,红孩儿,观音,西游记,天罡刀
+6531286584317182472_!_101_!_news_culture_!_佛家三年坐缸成肉身佛，肉身不腐且皮肤仍有弹性，毛发继续生长！_!_弥光法师,佛教,肉身不腐,坐缸,肉身佛
+6538669272304976387_!_101_!_news_culture_!_如来最大谎言是啥？一人早识破，还和观音演戏，孙悟空却蒙在鼓里_!_劝修功卷,小乘佛教,孔雀经,佛教,大乘佛法,唐僧,安邦天宝篆,观音,西游记,法华经,受生度亡经
+6474498306566258957_!_101_!_news_culture_!_如何评价《滕王阁序》的文学高度？_!_范仲淹,滕王阁序,王勃,滕王阁诗,闻一多,初唐四杰,岳阳楼记,张若虚,春江花月夜
+6539753454808597006_!_101_!_news_culture_!_这 10 部茅盾奖作品，改变了我的人生观_!_韩子奇,黄依依,黄雀记,秦腔,孙少平,芙蓉镇,白鹿原
+6543027763874365966_!_101_!_news_culture_!_中国男子代表油画作品《少女》竟卖了448万元，网友：美若天仙！_!_民族服饰,吴成伟,少女,惠安女,油画
+6500386380961546510_!_101_!_news_culture_!_西北人和东北人都有哪些差别？_!_文化,少数民族,内蒙古,游牧民族,东北人
+6470858730056450318_!_101_!_news_culture_!_因果报应真的存在吗？_!_于谦,了凡四训,马步芳,做善事,必然性,因果律,因果报应
+6552441619063243278_!_102_!_news_entertainment_!_高圆圆：出席活动，网友：好好的一条裙子，愣是整两块布吊起！_!_愣是整两,块布吊起,高圆圆
+6552286124008538632_!_102_!_news_entertainment_!_泰勒·斯威夫特出席活动，网友：穿得再普通也看得心血来潮_!_看得心血来潮,泰勒·斯威夫特,泰勒·斯威夫特出席活动
+6552359931213775107_!_102_!_news_entertainment_!_美剧有哪些常见的套路？_!_
+6552098106169623044_!_102_!_news_entertainment_!_创造101选手人设崩塌，露出本来面目？真应该和坤音四子好好学学_!_哈士奇,偶像练习生,朱天天,坤音四子,绝不认输,创造101
+6552308581046157831_!_102_!_news_entertainment_!_电影｜《天堂的孩子》：“不要冠军，只要第三名，第三名就好”_!_美好事物,运动鞋,天堂的孩子,剧情电影,马基德
+6552383857465754115_!_102_!_news_entertainment_!_《花木兰》预算超18亿人民币！传用《雷神》布景组_!_刘亦菲,真人版,花木兰,神奇女侠,雷神奇侠3 ：诸神黄昏,甄子丹,Ragnarok,女导演
+6552241775220621838_!_102_!_news_entertainment_!_张翰张钧甯同时公开恋情，甜蜜的不像话，网友：比郑爽古力娜扎配_!_流星雨,戏生情,张钧甯,张翰,古力娜扎
+6552371342010745348_!_102_!_news_entertainment_!_明星们的高考成绩，杨幂583分，周冬雨285分，还有一位考零蛋的！_!_神雕侠侣,高考,七月与安生,陈小春,关晓彤,南泥湾,红粉世家,笑傲江湖,高考成绩
+6552344478756110861_!_102_!_news_entertainment_!_快手短视频低俗遭整治后，想打翻身仗？网友：赞助跑男，极挑没用_!_快手,短视频平台,综艺,跑男
+6552282751150588424_!_102_!_news_entertainment_!_《忽而今夏》一部又羞耻又巨甜的青春剧，看看剧中有没有你的影子_!_五年高考三年模拟,肯德基,悄无声息,微微一笑很倾城,忽而今夏,美人为馅
+6552129356469109252_!_102_!_news_entertainment_!_如果当年张翰没有经历和郑爽的那段情感，如今他会发展的更好吗？_!_金粉世家,中国合伙人,千山暮雪,一起来看流星雨,命中注定我爱你,王耀庆,上古情歌,郑爽,全民目击,黄晓明,霸道总裁,花儿与少年,快乐大本营,莽荒纪,张翰
+6552345628804907524_!_102_!_news_entertainment_!_180506 别不承认！亭哥你是奶凶奶凶的boy没跑啦！_!_奶凶,白敬亭
+6552414638552646148_!_102_!_news_entertainment_!_王源：最近胖了，急！粉丝：露出了老阿姨般的微笑_!_小天使,哈哈哈哈哈哈哈哈哈哈哈哈,王源,粉丝
+6552856081687642631_!_102_!_news_entertainment_!_谢娜复出进军时尚界，然而衣品依旧辣眼睛？！是被张杰传染的吗？_!_谢娜,设计师,时装品牌,奚梦瑶,欢型
+6552266435291251204_!_102_!_news_entertainment_!_张柏芝出席活动，网友：这是我见过最长的裙子！_!_张柏芝出席活动,张柏芝
+6552294448858923528_!_102_!_news_entertainment_!_王琳出席红毯，网友：雪姨求你别再散发魅力了，这件裙子太赞了_!_雪姨,雪姨求,散发魅力,王琳出席红毯,裙子太赞
+6552457663781798403_!_102_!_news_entertainment_!_把女人当子宫，这绝不只是电影_!_使女,饲养人,女主人,惨无人道,妈妈桑,盲山,割礼
+6552114230395404804_!_102_!_news_entertainment_!_花60看一张照片？范丞丞的粉丝真是壕，范冰冰却骂弟弟白做嫁衣！_!_范丞丞,范冰冰,粉丝,艺人
+6552437171565888014_!_102_!_news_entertainment_!_贾乃亮一句“好久不见”让人心疼？_!_誓言,李小璐,此时此刻,贾乃亮,门事件,好久不见
+6552147349244739843_!_103_!_news_sports_!_山东鲁能有没有可能拿到今年的中超冠军？_!_
+6552260650813358350_!_103_!_news_sports_!_为什么国人能接受并喜欢石川佳纯，但不怎么喜欢张本智和？_!_
+6552405019419738375_!_103_!_news_sports_!_如何看待第一詹黑皮尔斯，公开赞美詹姆斯？_!_
+6552450452170998024_!_103_!_news_sports_!_如何看待帕托赛后在微博上霸气自封“天津之王”？_!_
+6552462121559392519_!_103_!_news_sports_!_2018乒乓球世锦赛为什么没有张继科？_!_
+6552411305020489998_!_103_!_news_sports_!_NBA如果用带队能力给现役球星排名，前五都有谁？_!_
+6552127754819600903_!_103_!_news_sports_!_2018斯诺克世锦赛半决赛 巫师希金斯率先挺进决赛_!_世锦赛,斯诺克,半决赛,威尔逊,希金斯
+6552277214946132228_!_103_!_news_sports_!_恒大爆冷客场惨败，上港中超两连败，你看好鲁能弯道超车时隔多年重夺榜首吗？_!_
+6552328842860560653_!_103_!_news_sports_!_刘伟解说乒乓球比赛的水平如何？_!_
+6415191599436792322_!_103_!_news_sports_!_他是菲尔普斯此生最大的冠军威胁，他的百万豪车更让“飞鱼”胆寒_!_奥迪A8L,A8L,孙杨,奥迪,奥迪A7
+6552314540632047879_!_103_!_news_sports_!_詹姆斯再次绝杀，3-0领先猛龙，你怎么看？_!_
+6552409566024303107_!_103_!_news_sports_!_库里差点被隆多故意垫脚？库里：对此我不予置评！_!_隆多,篮球,替补席,杜兰特,库里
+6551864669101883917_!_103_!_news_sports_!_因詹姆斯而成为历史背景的球队：一队14连败！唯他们让LBJ诚服！_!_凯尔特人,老鹰,詹姆斯,季后赛,步行者
+6552321302575710734_!_103_!_news_sports_!_恩比德绝望了，竟被这个新秀吊打啊！这家伙将和詹姆斯会师东决？_!_76人,霍福德,塔图姆,气势如虹,绿巨人
+6552059643433058829_!_103_!_news_sports_!_国足能否进入亚洲杯四强将与里皮续约一事直接挂钩_!_吉尔吉斯斯坦队,里皮,国足,亚洲杯,菲律宾队
+6552752801435353603_!_104_!_news_finance_!_上汽财务：这家主营车贷的公司2017年利润总额69亿 _!_上汽财务,金融业务,上汽集团,通用汽车,第一期,汽车金融,Wind,机器人
+6552351013104255245_!_104_!_news_finance_!_在银行ATM机取款，取款不成功，余额少了，钱没拿上，正确的解决办法是什么？_!_
+6552451536713155075_!_104_!_news_finance_!_天量货币存量形成解析——看好你的资本，放下你的杠杆_!_印钞,格林斯潘,对外投资,M2,金融资产
+6552299742645190916_!_104_!_news_finance_!_一家信用卡逾期五个月，他行信用卡会封卡降额吗？_!_
+6552480520012300813_!_104_!_news_finance_!_韩国央行：“预计中国将维持稳定房地产市场”_!_房地产,韩国银行,中国房地产市场,韩国央行,房地产市场
+6552461940478705924_!_104_!_news_finance_!_工龄40视同缴费15年实交25年，按5714元缴费，18年5月退休的退休金会是多少？_!_
+6552268852963574020_!_104_!_news_finance_!_MySQL最终有没有可能闭源？_!_
+6552120445863198989_!_104_!_news_finance_!_2018中国有哪些著名的营销咨询公司？_!_
+6552284483045818884_!_104_!_news_finance_!_「聚焦丝博会 展示新形象」“韩城制造”系列产品将亮相丝博会_!_国际博览会,韩城,丝博会,张新庄,丝绸之路
+6552401791873450509_!_104_!_news_finance_!_A股第一只独角兽5月8日上市，最新季度扣非利润同比下跌9%_!_Holdings,药明康德,私有化,IPO,上交所
+6552357185412661774_!_104_!_news_finance_!_马克思的影响从未离开——法国再掀马克思主义研究热潮_!_共产党宣言,资本主义,法兰西内战,政治经济学批判大纲,巴黎公社运动,巴黎,资本论,恩格斯,神圣家族,法国,1844年哲学和经济学手稿,德法年鉴,巴黎手稿
+6552430271789007112_!_104_!_news_finance_!_2018年，在建行存10万元，三年的利息是多少？_!_
+6552434756833772036_!_104_!_news_finance_!_苹果市值又涨了这么多 库克暗自发笑 暴涨原因竟然是他的出现_!_市值,伯克希尔,苹果手机,苹果,苹果公司
+6552429637236949507_!_104_!_news_finance_!_今年以来新三板累计发行股票金额超260亿元_!_股票,成交金额,新三板
+6552343401176498696_!_104_!_news_finance_!_用卡or玩卡！信用卡你最看重的是什么……_!_办理信用卡,信用卡,交通银行,象牙塔,交通银行信用卡
+6552244919451255303_!_104_!_news_finance_!_巴菲特：中国经济找到了秘诀，就像我们几个世纪前一样_!_经济体,伯克希尔,包括中国,内布拉斯加州,巴菲特
+6552449927123829251_!_104_!_news_finance_!_5月7日操作计划（附股）_!_特斯拉,高斯贝尔,创业板,沙河股份,广联达,株洲市
+6552173797942755591_!_104_!_news_finance_!_为什么中国的动画电影做不到美国的水平？_!_
+6552446466672034056_!_106_!_news_house_!_为什么郑州工资低，房价却居高不下？_!_
+6552337527452205576_!_106_!_news_house_!_专家：炒房客赔掉裤衩，只有傻子才帮忙接盘！_!_炒房客,楼市,买房,楼盘,房价
+6552280882227446276_!_106_!_news_house_!_最新18个城市房价数据 青岛17024元/㎡ 又涨了！_!_青岛,房价,郑东新区,上月,楼盘
+6552440531652182276_!_106_!_news_house_!_2018年，哪些城市的房子可以买？_!_
+6552200782282752264_!_106_!_news_house_!_海南买房先签订购房合同，挂靠公司交两年社保，满两年后再过户，这样做靠谱吗？_!_
+6552400255273402631_!_106_!_news_house_!_常州房价还会上涨吗？_!_
+6552235105354514702_!_106_!_news_house_!_如何看待央视财经专家马光远的炒房无罪论？_!_
+6552297671820837124_!_106_!_news_house_!_运动会都有什么项目？_!_
+6552415240649179400_!_106_!_news_house_!_武汉房价未来会如何发展？_!_
+6552140216449106179_!_106_!_news_house_!_资管新规后，房价还会继续上涨吗？_!_
+6552238845293756936_!_106_!_news_house_!_司法拍卖 中海瓦尔登湖 北京法拍房2万/平米在售 可贷款 可垫资_!_
+6552297292018221571_!_106_!_news_house_!_浙江阿姨海南购房记：限购让大家竹篮打水一场空_!_中介,海南,二手房,房东,海南省
+6552321469488038148_!_106_!_news_house_!_小产权房与安置房的区别？_!_
+6552274446541914628_!_106_!_news_house_!_未取得房本的安置房，购买需慎重！_!_交房,被拆迁人,安置房,买受人,买卖合同
+6552294641083875854_!_106_!_news_house_!_物业的好坏决定楼盘的“软实力”_!_物业服务,物业公司,物业费,物业,物业管理
+6552457256355496206_!_106_!_news_house_!_2018年开始房地产又要暴涨了吗？_!_
+6552375527871611406_!_106_!_news_house_!_中海绿城华润置地共30强房企对垒杭州亚运村 竞争超出你想象_!_华润集团,绿城,华润置地,宋卫平,开发建设
+6552487891140870663_!_107_!_news_car_!_汽车油表灯亮了，还能跑多远？90%不知道_!_加油站
+6552347142537609731_!_107_!_news_car_!_外观抄袭内饰抄袭也算了，这台国产车连名字都要抄袭_!_昌河,CS55,昌河A6,国产车,A6,北汽绅宝
+6552230833694966276_!_107_!_news_car_!_出门在外担心的就是手机没电，多功能车载充电器，出行无忧更畅快_!_充电器,车载充电器,点烟器,多功能,充电器多功能
+6552413285004608014_!_107_!_news_car_!_乱清积碳才真伤车！老司机教你一招！比去4s店管用多了_!_润滑剂,告别高温,行车记录仪,积碳,汽车腰靠
+6552471015425835524_!_107_!_news_car_!_大众与滴滴合资；氢能源汽车产业平台成立；人人车融资3亿美元_!_氢能源,众泰汽车,Auto,CATL,滴滴出行,合资公司,白名单,戴姆勒
+6552348690558747139_!_107_!_news_car_!_这货再令国人疯狂，吉利新车标比路虎惊艳，仅9万奥迪见着绕路走_!_吉利,宝腾,马来西亚,商用车,东南亚,普拉多
+6552296023262232836_!_107_!_news_car_!_高速上保持安全车距，就一定不会发生追尾事故吗？_!_
+6552404016582623747_!_107_!_news_car_!_为什么越是有钱人越是选择分期付款购车，而穷人却要全款买？_!_买车,银行,穷人,分期付款
+6552389224232911374_!_107_!_news_car_!_进口大众T6，休闲商务惬意畅行_!_进口大众t6前舱展示
+6552287429380801037_!_107_!_news_car_!_清凉一夏，推荐几款精致冰丝汽车坐垫，让你爱车瞬间档次直线飙升_!_越野车,汽车坐垫,冰丝汽车坐垫
+6552414176948519437_!_107_!_news_car_!_四月车企销量首批公布，观致复活，同比增长400%！日韩势头凶猛_!_经销商,观致汽车,起亚销量,长安马自达,马自达
+6552368584696267277_!_107_!_news_car_!_众泰又出手了！最新SUV伪装车曝光，外形酷似路虎揽胜_!_路虎揽胜,R21,众泰
+6552463699091652867_!_107_!_news_car_!_宝骏560这车怎么样？_!_
+6552337520233808397_!_107_!_news_car_!_出门带雨伞太不方便，老婆包里装了个小球球，雨天瞬变透明雨衣_!_晴雨挡,遥控折叠无人机,放大器,小便器,手机屏幕
+6552357382532366855_!_107_!_news_car_!_汽车行业2018年2月产销数据_!_中国汽车工业协会,产销,半挂牵引车,企业销量,汽车行业
+6552378386138792205_!_107_!_news_car_!_皮卡车柴油和汽油哪个好？_!_
+6552384036348625156_!_107_!_news_car_!_吉利车外观普通，内饰没特色，油耗不低，为什么网上说销售量还很高？_!_
+6552334274169667844_!_107_!_news_car_!_选车牌号有哪些禁忌？_!_
+6552374202760954372_!_107_!_news_car_!_在美国花一百美元跟在中国花一百人民币一样吗？_!_哈根达斯,购买力,美国花,奢侈品,人民币
+6552356913072308750_!_108_!_news_edu_!_于都县葛坳乡中心小学，快来认领母校_!_于都县,中心小学,葛坳乡
+6552453078690300423_!_108_!_news_edu_!_西安工程大学9人获批2017年陕西省三个高层次人才项目_!_陕西省,人才项目,千人计划,中共陕西省委,特支计划
+6552271287413113357_!_108_!_news_edu_!_武汉大学2018年录取分数线 历年录取最高最低分_!_高考志愿,大数据,广西高考志愿,省控线,武汉大学
+6552477189114167811_!_108_!_news_edu_!_即便考研三次，也要进入北大，成功的可能性有多大？_!_北大,北京,考研
+6552346686641930756_!_108_!_news_edu_!_莒县碁山二小节假日德育作业彰显人文特色_!_金点子,碁山镇,义务教育,义务教育法,焦点访谈,新闻联播,辍学率,司法员
+6552472413559325192_!_108_!_news_edu_!_从元谋人到北京人，再到山顶洞人_!_北京猿人,尼安德特人,山顶洞人,元谋人,直立人
+6552336863464521998_!_108_!_news_edu_!_衡水中学的体育课都干什么？_!_
+6552477286325551623_!_108_!_news_edu_!_高校人才工作改革——破局之举还是无奈挣扎？_!_该校,权力下放,自主权,武汉大学,引进人才
+6552292016582033671_!_108_!_news_edu_!_流浪狗扰乱正常教学秩序，破坏学生们生活，该怎么办？_!_
+6552302909814472963_!_108_!_news_edu_!_吉林大学为什么力压大工和哈工大，成为东北第一学府？_!_
+6552318534054052356_!_108_!_news_edu_!_中国五所大学5月共迎校庆 赏百年高校美景_!_北京大学,国立同济大学校史,南京大学,校庆,同济大学,复旦大学
+6552305755888812547_!_108_!_news_edu_!_琅琊新闻网科技小记者首次亮相 参与机器人竞赛直播_!_直播,新闻网,琅琊新闻网,机器人
+6552348083630375182_!_108_!_news_edu_!_你在辅导孩子数学作业时，孩子说你跟老师讲的不一样，你是怎样回答的？_!_
+6552366544620356109_!_108_!_news_edu_!_淄博信息工程学校职业教育活动周教学成果展示_!_启动仪式,淄博信息工程学校,无人机,张店区,淄博市
+6552277376481362180_!_108_!_news_edu_!_你见过最努力的人现在都变成什么样子了？_!_
+6552268397525074180_!_108_!_news_edu_!_事业单位的在职人员，可以再报考事业单位吗？_!_
+6552118280683782669_!_108_!_news_edu_!_北流市永顺小学：榜样引领，做一个有魅力的人！_!_十万个为什么,学习标兵,中华上下五千年,李欣怡,魅力学子,童话故事,魅力教师,优秀,昆虫记,永顺小学,北流文艺,爱的教育,钟辉,自然与科学
+6552143691262198020_!_108_!_news_edu_!_高考分数在二本中段线，想报临床医学专业，有希望吗？_!_
+6552101611253006856_!_108_!_news_edu_!_武汉外校“00后”高中生唱革命歌曲_!_革命歌曲,00后,十送红军,艺术教育,我和我的祖国,高中生,黄河大合唱,保卫黄河
+6552156830754144519_!_109_!_news_tech_!_几乎每一个银行工作人员都劝说顾客，不要把钱存入余额宝，对此你怎么看？_!_
+6552462937259246093_!_109_!_news_tech_!_TOPPERS智能AI音箱问世：告别固定电话，颠覆移动通信行业！_!_人工智能,AI,音箱,智能家居,音响
+6552375825352622343_!_109_!_news_tech_!_粉丝每天稳定增长产品资讯获得更多曝光，如何利用免费流量？_!_
+6552360664822710788_!_109_!_news_tech_!_像极了谷歌Chorme的Microsoft Edge浏览器_!_Windows 10,Firefox,Safari,微软,Chrome
+6552374028911247624_!_109_!_news_tech_!_想注册滴滴，驾龄不满3年怎么办？_!_
+6552496126996513031_!_109_!_news_tech_!_作为一个5s 16G用户，iphone8plus和iphone7plus买哪个比较好？_!_
+6552492650727997700_!_109_!_news_tech_!_SEO自学能学会吗？_!_
+6552422669810139406_!_109_!_news_tech_!_如何删除手机里没用的文件？_!_
+6552346279127548173_!_109_!_news_tech_!_现在有什么好的手机自动挖矿软件？_!_
+6552464238206517507_!_109_!_news_tech_!_F-Secure破解酒店智能锁，打开数百万个酒店房间能让锁具大洗牌吗？_!_
+6552376509112254989_!_109_!_news_tech_!_8场高端对话 50余场专业论坛邀你同台论道_!_大数据,数字经济,高端对话,数据安全,意法半导体
+6552299457491239432_!_109_!_news_tech_!_第一季度全球销量排行，小米华为紧追三星苹果！_!_小米,OPPO,三星,智能手机,华为
+6552388635369406734_!_109_!_news_tech_!_小米手机扩容之后有什么不一样的体验？_!_
+6552453081798279427_!_109_!_news_tech_!_贴片元件用什么电烙铁头好？_!_
+6552293352568520963_!_109_!_news_tech_!_你平时网购吗，都用什么平台网购呢？_!_
+6552131788502729220_!_109_!_news_tech_!_腾讯没有梦想_!_腾讯微博,微视,任宇昕,腾讯,短视频,腾讯视频,张小龙,马化腾,旅行青蛙,刘炽平,腾讯传,SNG,光荣使命,恋与制作人
+6552373226251485709_!_109_!_news_tech_!_Top500发起人将领衔ASC18超算与AI国际论坛_!_人工智能,高性能计算,超算,超级计算机,南昌大学
+6552438752562643204_!_109_!_news_tech_!_如何用手机号码定位？_!_
+6552330912724418823_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552145819175223566_!_110_!_news_military_!_如果发生战争，军事专家能不能成为军事指挥员？_!_
+6552134853612011779_!_110_!_news_military_!_真实战场上迫击炮兵的生存率如何？_!_
+6552151490759229703_!_110_!_news_military_!_M4A1跟M16的差别在哪里？_!_
+6552256485441667342_!_110_!_news_military_!_如何看待美国将长时间停止对"白头盔"组织的资金供应？_!_
+6552281299522945294_!_110_!_news_military_!_以色列警告称如果战机被击落将会轰炸俄军事基地，你怎么看？_!_
+6552471659926782215_!_110_!_news_military_!_中国从古到今谁才是最有军事才华的？_!_
+6552384395964055811_!_110_!_news_military_!_印度光辉战机真的有那么不堪吗？_!_
+6552453338040893955_!_110_!_news_military_!_都错了！美官方证实战机坠毁真相，要求盟友必须给出回应_!_战机,运输机,坠毁,侦察机,美联社
+6552446094326890755_!_110_!_news_military_!_为何美国为首的西方国家不允许伊朗研发核武器，但对以色列拥核却视而不见？_!_
+6550085272908857613_!_110_!_news_military_!_如果日本派一个师团登陆夏威夷而不是空袭珍珠港，太平洋舰队会不会覆灭？_!_
+6552159923541639427_!_110_!_news_military_!_女乒今天排兵布阵不合理，丁宁昨天刚打硬仗，今天应该打第一单打，你认为呢？_!_
+6552485872816620046_!_110_!_news_military_!_汶川地震第一个跳伞营救的勇士，他叫李振波，现在怎么样了_!_李振波,汶川地震,跳伞,泥石流,空降兵
+6551768593825005838_!_110_!_news_military_!_反导系统在作战中的作用大不大？_!_
+6512026669954367751_!_110_!_news_military_!_航母甲板多人一字排开走是在干什么？_!_航母,航母甲板,甲板,飞行甲板,捡垃圾
+6552129580797264392_!_110_!_news_military_!_传承红大、抗大血脉，让三尺讲台连接未来战场_!_国防大学,陈晖,部队,航空兵,联合作战
+6552326740163691015_!_112_!_news_travel_!_这次夏威夷真的很火，真相令人是喜是忧呢？！_!_夏威夷,夏威夷群岛,热带雨林,基拉韦厄火山,夏威夷岛
+6552468792960614919_!_112_!_news_travel_!_国内这5个景点最黑，都是热门景点，即使再有钱也不要去了！_!_内地人,倚天屠龙记,布达拉宫,少林寺,阳朔西街,峨眉山
+6552486858847158797_!_112_!_news_travel_!_还在看抖音，羡慕别人喝摔碗酒？西安好玩的景点现在就告诉你！_!_大雁塔,全国重点文物保护单位,骊山,华清池,回民街
+6552377320559083780_!_112_!_news_travel_!_去西藏旅游，佳能5D4配什么镜头好？_!_
+6552253156900733191_!_112_!_news_travel_!_请问去泰国旅游笔记本电脑在异地能上网能用吗？_!_
+6552141691346747656_!_112_!_news_travel_!_川藏线旅游包车费用是如何算的？_!_
+6552292118629450254_!_112_!_news_travel_!_第一个对全国公安民警免费开放的国家5A级景区、中华十大名山长白山五一小长假迎来首批民警游客_!_惠民五条,长白山公安局,董海军,长白山景区,人民警察
+6552241275737735428_!_112_!_news_travel_!_国内哪一家航空公司的飞机餐最好吃？_!_
+6552334266951270670_!_112_!_news_travel_!_当您有钱了以后最想跟谁出去旅行？_!_
+6552350894321566221_!_112_!_news_travel_!_情侣退房，保洁阿姨进门清扫时傻眼了，保洁阿姨：不由得哭了！_!_保洁阿姨,退房,情侣,阿姨
+6552465077197340936_!_112_!_news_travel_!_河北省保定市易县和定兴，将来哪里发展比较好？_!_
+6551991829359772173_!_112_!_news_travel_!_群发短信在旅游行业的应用_!_旅行社,群发短信,机票,旅游行业
+6552487284275413252_!_112_!_news_travel_!_说到出国旅游，人们第一个想到的是哪几个国家？_!_
+6552484533629878788_!_112_!_news_travel_!_巍巍太行 世外桃源-薄壁、双底、上云台-南太行自驾路书（1）_!_双底村,武家湾村,太行山,农家乐,薄壁镇
+6552168436779188743_!_112_!_news_travel_!_恩施大峡谷风情小镇_!_小镇,观景台,恩施大峡谷
+6552291296722027011_!_112_!_news_travel_!_福州将积极扶持乡村旅游发展_!_福州,福清,连江,乡村旅游,长乐
+6552443390678532360_!_112_!_news_travel_!_书上说，地球上大量的雨水蓄满了火山口，形成海洋。雨水是淡水，海水为什么是咸的？_!_
+6552429617322394115_!_112_!_news_travel_!_夕阳下的日照东夷小镇花田好美_!_日照市
+6552398778198917379_!_113_!_news_world_!_侣行在伊拉克，叙利亚等的土豪行为，为后来的探险家造成巨大的伤害，您怎么看？_!_
+6552400724368556558_!_113_!_news_world_!_英外交大臣启程赴美 劝特朗普勿“毁”伊核协议_!_英国外交大臣,特朗普,特蕾莎,伊朗,鲍里斯
+6552464317202039044_!_113_!_news_world_!_同样是石油储量多的国家，为什么伊朗和沙特的发展区别这么大？_!_
+6552381350773522696_!_113_!_news_world_!_向小布什扔鞋的那位伊拉克记者，后来被怎么处理了？_!_
+6552427231979766280_!_113_!_news_world_!_弱国无外交！盘点史上最无耻的六个战争借口_!_伊拉克,战争借口,缅因号,驱逐舰,北部湾决议案,参考消息,叙利亚
+6552142897490166029_!_113_!_news_world_!_美国建国不过两百年，为什么会成为世界强国？_!_
+6552493808100049155_!_113_!_news_world_!_如果现在是希拉里当总统，美国的外交政策会是什么样？_!_
+6552449777831772423_!_113_!_news_world_!_美国参议员麦凯恩说他的葬礼上不请特朗普，大家怎么看？_!_
+6552290543240479239_!_113_!_news_world_!_女子腿毛比男生还要长，因此找不到男友失去信心，网友纷纷鼓励_!_
+6552469320855716356_!_113_!_news_world_!_战场上俄军充满激情？接到指令后众多俄军辞职，事情没有那么简单_!_俄罗斯,俄军,叙利亚
+6552306916247208206_!_113_!_news_world_!_为什么罗斯福新政时宁愿把过剩的牛奶、棉花扔掉也不免费分发给穷人？_!_
+6552327336031682820_!_113_!_news_world_!_为什么美国人一到18岁，马上会被父母赶出家门？_!_
+6552268187889566212_!_113_!_news_world_!_男子在家中浴池洗澡突然一声大叫，女友进门就报警，真相让人无语_!_男子
+6552306108801745415_!_113_!_news_world_!_大熊猫在各国现状：日本的窒息死亡 德国的喝酒猝死 泰国令人心酸_!_嫣嫣,林慧,兴兴,大熊猫,柏林动物园
+6552424438317449479_!_113_!_news_world_!_最近有坐乌克兰航空的吗？安全吗？_!_
+6552471629006373384_!_113_!_news_world_!_不顾联合国强烈反对！以色列战机悍然轰炸核电站，中东惨剧发生了_!_以色列空军,以色列,反应堆,核电站,伊拉克
+6552373823935611405_!_113_!_news_world_!_女子在大街上穿渔网裙被路人围观，知道原因纷纷拍案叫绝_!_
+6552467736163779085_!_115_!_news_agriculture_!_农村租地可要小心啦！这几种情况租出去可能就要后悔，当心了_!_农村,租地,农民朋友,可耕地
+6552342765252903428_!_115_!_news_agriculture_!_昨天7毛，今天才卖4毛，蒜薹一天一个收购价，蒜农仍能坦然面对_!_收购点,蒜农,蒜薹
+6552247253002617357_!_115_!_news_agriculture_!_将贫困户融入市场主体之中 恩施特色产业带动62.5万人脱贫_!_恩施州,汪营镇,121 X,合作社,特色产业
+6552348093315023108_!_115_!_news_agriculture_!_2018年种植的苹果的市场行情怎么样？_!_
+6552455273947070984_!_115_!_news_agriculture_!_大兴安岭垦区做强“五彩产业”_!_倒春寒,中药材,农业生产,除草剂,大兴安岭
+6552255831763583501_!_115_!_news_agriculture_!_福州金鱼亮相成都农博会 今年计划举办30场巡展_!_福州金鱼,福州市,农博会,成都,福州
+6552406423601414659_!_115_!_news_agriculture_!_小记——老张的故事_!_老家人,同龄人,老张,自行车,好日子
+6552312788436386318_!_115_!_news_agriculture_!_最适合懒人种的8种果苗盆栽，好养易活，营养丰富越吃越健康_!_大红袍,盆栽,树苗,杨梅树,猕猴桃
+6552313813121630467_!_115_!_news_agriculture_!_为什么现在很多农户都不愿意养猪？_!_
+6552288459837407751_!_115_!_news_agriculture_!_湖南宁乡：产粮大县的田野“四则运算”_!_宁乡,陈德奇,紫云英,农业,有机肥
+6552385099642438158_!_115_!_news_agriculture_!_农村种植生姜，技术员讲的这些科学管理方法很管用！_!_技术员,科学管理,生长期,生姜,姜辣素
+6552443714344583688_!_115_!_news_agriculture_!_农村老话“穷柴火，富水缸”，你知道是什么意思么？_!_农村
+6552258024189198595_!_115_!_news_agriculture_!_冬瓜子如何育苗？_!_
+6551552232481882381_!_115_!_news_agriculture_!_红豆杉能插栽吗？_!_
+6552353604819223047_!_115_!_news_agriculture_!_猪价持续下跌，养猪赔钱！二师兄：俺也不想那么贱_!_大北农,猪周期,蒜你狠,养猪人,养猪
+6552302730231153156_!_115_!_news_agriculture_!_还养花种草呢？别傻了，现在都种果苗盆栽，好看又好吃，贼有情调_!_调养作用龙眼苗,山竹,杨梅属,果树苗,杨梅科
+6552414159005286926_!_115_!_news_agriculture_!_今年开始，这个证农村也能办了，办好的农民要放好，能换小十万！_!_农人,不动产证,宅基地,农村,自建房
+6552236904538964228_!_115_!_news_agriculture_!_什么品种的玉米种最抗倒伏？_!_
+6552369182099374599_!_116_!_news_game_!_韩国3大网红终极对决COS《守望先锋》D.va谁更美_!_D.Va,韩国,美少女,COSPLAY,Yurisa,守望先锋
+6552479584493764872_!_116_!_news_game_!_世乒赛决赛第二盘，樊振东3:0取胜削球手菲鲁斯，怎么评价本场比赛？_!_
+6552089643305665038_!_116_!_news_game_!_SUP三战全胜领跑B组 日本冠军一胜难求_!_SUP,Buff,英雄联盟,PGM,巨魔,Triple
+6552489139210551812_!_116_!_news_game_!_斗鱼黄金大奖赛——国内最强吃鸡赛事的养成记！_!_FPP,Numb,TPP,斗鱼,绝地求生,黄金大奖赛
+6552249091982622989_!_116_!_news_game_!_你认为红白机历史上最难的三款游戏分别是什么？_!_
+6552474744845762819_!_116_!_news_game_!_《荒野行动》的鼠标灵敏度要怎么调才最好？_!_
+6552337937558667534_!_116_!_news_game_!_你玩过最老的游戏中，哪些已经关服了？哪些又复活了？_!_
+6552386695742882052_!_116_!_news_game_!_你玩过剧情最好的RPG游戏是什么？_!_
+6552323004875931918_!_116_!_news_game_!_《王者荣耀》里的排位赛如果可以禁三样装备，你会禁什么？_!_
+6552447355906425357_!_116_!_news_game_!_王者荣耀：李白原画重做多次，皮肤炫酷，网友：越帅越坑？_!_酒剑仙,王者荣耀,英雄战迹,原画,李白
+6552407332305764877_!_116_!_news_game_!_绝地求生：玩家这操作被称为最强98K！网友：1千块皮肤买的值！_!_狙击枪,AWM,98k,枪械皮肤,98K
+6552273403804385544_!_116_!_news_game_!_Steam上有哪些适合用手柄玩的游戏？_!_
+6552269708609978894_!_116_!_news_game_!_QGhappy最后的倔强！使用夺冠阵容连输三局后将要面临保级_!_QGhappy,保级圈,打野,Alan,季后赛
+6552454883075686920_!_116_!_news_game_!_DNF：红狗养成之路第四十六天_!_dnf
+6552422678504931848_!_116_!_news_game_!_不请明星的吞鲲游戏，凭什么比肩贪玩蓝月？_!_吞鲲系列游戏,浏览器,网页游戏,蓝月,游戏
+6552474439395574023_!_116_!_news_game_!_YTG的路西法怎么又成替补了？_!_
+6552345030206423299_!_116_!_news_game_!_《刺激战场》27杀吃鸡很厉害吗，你怎么看？_!_
+6552294697698590989_!_116_!_news_game_!_ns破解方法如果被公开，对正版玩家有什么影响？_!_
+6552329565551722765_!_116_!_news_game_!_如果在《王者荣耀》里选一个英雄陪你在荒岛生存一辈子，你会选谁？_!_
+6552128011842355720_!_116_!_news_game_!_吃鸡游戏手机精选，这几款手机玩手游都不错_!_小米MIX,吃鸡,吃鸡手机,海思麒麟,P20
+6552019389271507470_!_101_!_news_culture_!_让·杜布菲：作品如同儿童涂鸦，却被誉为“最杰出法国画家”_!_杜布菲,抽象画,艺术流派,涩艺术,佛系
+6552082798247674382_!_101_!_news_culture_!_赖晓伟讲述：补天石惊现人间，红楼梦要换主人？_!_山海经·大荒北经,长白山,补天石,大荒山,长白山江岗志略,红楼梦,女娲氏
+6552278341984977160_!_101_!_news_culture_!_上联：一度春风一度愁，黄花落尽，如何对下联？_!_
+6552464760288313608_!_101_!_news_culture_!_上联：酒醉不知日暮白，如何对下联？_!_
+6552250293290008845_!_101_!_news_culture_!_上联：一夜春风绿两岸，下联怎么对？_!_
+6552491621873615118_!_101_!_news_culture_!_上联：天高任鸟飞，下联如何对？_!_
+6552344564936474883_!_101_!_news_culture_!_古风不老山河壮，如何接下句？_!_
+6552236974516732424_!_101_!_news_culture_!_诗词里的林州太行大峡谷_!_何巍巍,大峡谷,林州,太行山,苦寒行
+6552345908783087879_!_101_!_news_culture_!_上联：春夏秋冬交好运，如何对下联？_!_
+6552128040200044804_!_101_!_news_culture_!_上句：浊水三江一夜雨，如何接句？_!_
+6552327991391682824_!_101_!_news_culture_!_劳动能否增加智慧？_!_
+6552429857890894087_!_101_!_news_culture_!_上联：金龙盘玉柱，怎么对下联？_!_
+6552383391382110468_!_101_!_news_culture_!_特朗普跟里根有可比性吗？_!_
+6552131511024353543_!_101_!_news_culture_!_会龙二字作一对联，最好将二字分别放在上下联的头一字，如何作？_!_
+6552320675049112071_!_101_!_news_culture_!_为什么学书法的人大多长寿？让科学告诉你！_!_胰岛β细胞,病由心生,书法家,不良情绪,书法养生论,中国康复,辛德勒,书法,交感神经,太极书法,调节情绪,红岩,当代书法十一家遗作展
+6552280874774167822_!_101_!_news_culture_!_有人说南方口音保留有上古音，那么粤语是否跟南越国有关呢？_!_
+6552438765351076100_!_101_!_news_culture_!_上联，东启明西长更，南箕北斗，如何对下联？_!_
+6552337976062378504_!_101_!_news_culture_!_成都画院纪念5.12汶川地震十周年全国书画名家赈灾作品回顾展开幕_!_美术馆,成都画院数字流动美术馆,汶川,艺术家,特大地震映秀震中纪念馆
+6552320214141239566_!_101_!_news_culture_!_佛罗伦萨画派著名艺术家波提切利的代表作有哪些？_!_
+6552321673930998030_!_101_!_news_culture_!_宋江身上常带着大批金银，他哪来的那么多钱财？_!_
+6551989548035867149_!_102_!_news_entertainment_!_《人类清除计划 》三部曲，血洗全城的复仇之夜已拉响警报_!_人类清除计划,Dolby,英语原声,Digital,DTS
+6552344278209659405_!_102_!_news_entertainment_!_最美“杨贵妃”林芳兵，年过半百依旧优雅，儿子样貌像极了她_!_金鹰奖,唐明皇,杨玉环,夜间行车,杨贵妃,幽谷恋歌,林芳兵,大众电影百花奖,燕子李三,白发魔女
+6552347431567098372_!_102_!_news_entertainment_!_泪眼问花花不语 春去春回情何依_!_我想有个家,赵丽颖,任松,许超,汤河
+6552272607616434696_!_102_!_news_entertainment_!_闪婚不是因为拜金 大S曝一见钟情汪小菲内幕_!_汪小菲,大s,摄影棚,真人秀,大S
+6552382220256936456_!_102_!_news_entertainment_!_可爱风的范冰冰是这个样子的！像极了偷吃的咘咘_!_蒋欣,咘咘,妈妈是超人,萌萌哒,范冰冰
+6552397368363319812_!_102_!_news_entertainment_!_看了菅纫姿小姐姐吃的饭菜，再也没有当明星的梦想了，网友：心疼_!_猎场,演艺圈,女明星,女演员,娱乐圈,罗伊
+6552236533598913032_!_102_!_news_entertainment_!_范冰冰穿军装，刘亦菲穿军装，孙俪穿军装，看到最后一位笑喷了_!_刘亦菲,真人秀,范冰冰
+6552467166266917384_!_102_!_news_entertainment_!_谢娜复出，几次在舞台上哽咽 同时也祝她生日快乐_!_陈乔恩,快乐大本营,杜海涛,谢娜,蔡少芬,大本营
+6552398450581832195_!_102_!_news_entertainment_!_扎心！同父母所生颜值差异大的星二代，有的子承父业，有的丢尽颜面_!_陈飞宇,曾志伟,向佑,曾国猷,刘德盛,七月与安生
+6552296311691936004_!_102_!_news_entertainment_!_明星看自己演的电影、电视剧是什么感觉？_!_
+6552311008604455437_!_102_!_news_entertainment_!_这些明星简直要占据当代小鲜肉半壁江山呐，白敬亭，费启鸣，等等_!_
+6552360176832217607_!_102_!_news_entertainment_!_刚求婚女友就曝出怀孕 赌王儿子何猷启进展太神速_!_赌王,何鸿燊,双喜临门,Gigi,何猷启
+6552476370423775748_!_102_!_news_entertainment_!_TO:Big Bang，你们陪我走过好长一段青春_!_戏里戏外,爱情公寓,生活大爆炸,谢耳朵,第十一,Amy
+6552293471024054787_!_102_!_news_entertainment_!_被一部电影毁了名声的5位明星，王宝强姜文上榜，第一至今没翻身_!_中国电影,分手大师,王宝强姜文上榜,大闹天竺,无极,陈凯歌,一步之遥,霸王别姬,电影毁,让子弹飞,刘若英
+6552450249888104974_!_102_!_news_entertainment_!_霍建华夫妇亲热，网友：老霍的表情亮了_!_林心如,粉丝,小白菜,夫妻二人,霍建华
+6552249864887992845_!_102_!_news_entertainment_!_赵丽颖出道十二年，自拍却都是一个样_!_女明星,剪刀手,赵丽颖,嘟嘴,自拍照
+6552311473396253198_!_102_!_news_entertainment_!_《西游记》女儿国国王的美你喜欢哪一位？_!_
+6552033221771526664_!_102_!_news_entertainment_!_佟丽娅的身材真是绝了，充满青春活力如少女，没人敢跟她拼体重！_!_牛仔裤,陈思诚,杨幂,美丫,宫锁心玉,佟丽娅
+6552461489360339470_!_103_!_news_sports_!_马龙暖心庆祝萌翻全场！国乒两将6-0吊打德国！国乒即将夺九连冠_!_2-0领先,马龙,樊振东,德国,国乒
+6552281400395956744_!_103_!_news_sports_!_曼联穆帅爱将自称值5000万镑？英媒：一半都达不到，不配效力曼联_!_卡拉格,爆炸头,曼联,利物浦,费莱尼
+6552304329762537732_!_103_!_news_sports_!_詹姆斯再次绝杀，3-0领先猛龙，你怎么看？_!_
+6552104232726561294_!_103_!_news_sports_!_中国姑娘破世界纪录 该项目比马拉松更辛苦 选手曾一场比赛瘦8斤_!_司天峰,竞走,马拉松,尹航,中国姑娘
+6552404528476455428_!_103_!_news_sports_!_LCX潮：最帅篮球鞋 OFF-WHITE x Nike Hyperdunk HD2017 OW评测_!_篮球鞋,FOAM,Hyperdunk,Booker,Nike
+6552421779803996430_!_103_!_news_sports_!_你如何看待广州恒大客场0比3完败大连一方？_!_
+6552316657899946243_!_103_!_news_sports_!_骑士vs猛龙G3，詹姆斯偷听猛龙战术，被发现后现机智反应，你怎么看？_!_
+6552340243993854211_!_103_!_news_sports_!_2019年亚洲杯分组抽签揭晓，中国队首夺亚洲杯的机会真的来了吗？_!_
+6552315341723140360_!_103_!_news_sports_!_为什么10年总决赛巅峰詹姆斯韦德波斯三巨头输给司机的小牛？_!_
+6552442279406076424_!_103_!_news_sports_!_詹姆斯谈最后一攻：乐福发球时只想找我_!_詹姆斯,乔治-希尔
+6552250635125785102_!_103_!_news_sports_!_2串1精析：马拉加主场保平争胜 三箭击穿胜利船_!_广岛三箭,星期日,波多尔斯基,马拉加,近况简析
+6552315165570760964_!_103_!_news_sports_!_詹姆斯再一次绝杀了，猛龙输球，你怎么看？_!_
+6552413164267372814_!_103_!_news_sports_!_离开新疆的李秋平，下赛季会到哪个球队执教？_!_
+6552277485604569603_!_103_!_news_sports_!_骑士主场战猛龙詹姆斯再创伟大纪录，今天詹皇给猛龙上了一课_!_猛龙,德罗赞,詹姆斯,骑士,季后赛
+6552252193011925507_!_103_!_news_sports_!_亚洲杯，国足抽了个好签？_!_亚洲杯,阿联酋,亚洲杯赛,国足
+6552313456161194247_!_103_!_news_sports_!_巴西足球队历史最强后卫有哪些？_!_
+6552249798760595972_!_103_!_news_sports_!_霍福德定胜负，凯尔特人客胜76人彩带抢镜，史蒂文斯太强了_!_霍福德,贝里内利,凯尔特人,76人,绿衫军
+6552372136344813837_!_103_!_news_sports_!_凯尔特人3-0了，但为什么赛前ESPN21位专家有18位都预测76人晋级？_!_
+6552276016939663624_!_104_!_news_finance_!_孙正义没有当过亚洲首富吗，为什么？_!_
+6552364445438312963_!_104_!_news_finance_!_瑞郎币FRCoin：金融王国走出的新虚拟货币，规避风险的财富摇篮_!_规避风险,虚拟货币,瑞郎币FRCoin,瑞士,瑞士银行
+6552461242190004483_!_104_!_news_finance_!_如何使托管班得到长久持续的发展？_!_
+6552440813350027783_!_104_!_news_finance_!_农业A股上市企业净利润50强排行榜_!_北大荒,新希望,牧原股份,农业行业,东方集团
+6552253391874031886_!_104_!_news_finance_!_原来每周发新股十家以上，现在是两家，股市因此就好了吗？_!_
+6552423671347020295_!_104_!_news_finance_!_信用卡马上要逾期了怎么办？怎么做不会影响征信？_!_农业银行,建设银行,中国银行,还款宽限期,工商银行,还款,平安银行,持卡人,信用卡,中信银行,逾期,浦发银行,交通银行
+6552474086675579140_!_104_!_news_finance_!_50岁了，一直是电商方面自由职业者，现在准备职业炒股从哪里入手？_!_
+6552276926164107790_!_104_!_news_finance_!_苌楚解盘：5.6-5.7黄金全线看涨原油再次攻顶周一展望附解套_!_解套,MA10,布林带,石油,MA5,原油,委内瑞拉,MACD
+6552348722221548036_!_104_!_news_finance_!_2018年A股百亿富豪：何享健成首富 贾跃亭退出榜单_!_百亿,王卫,周群飞,2018年A股百亿富豪榜,周鸿祎,何享健
+6552435965858677262_!_104_!_news_finance_!_赚得盆满钵满不知足，这国趁机抬价，中国一出手，500亿化为乌有_!_俄罗斯,原油,中石化,进口量,基准价
+6552382576768582152_!_104_!_news_finance_!_巴菲特强烈警告：这种投资，会死得很惨！_!_零和游戏,虚拟货币,CNBC,伯克希尔,中央银行,巴菲特,台积电,比特币,苹果,伯克希尔哈撒韦
+6552496955686126094_!_104_!_news_finance_!_股神大赞苹果，比特币毫无价值？_!_富国银行,伯克希尔·哈撒韦,苹果公司,比特币,巴菲特,数字货币,伯克希尔哈撒韦
+6552130846462050823_!_104_!_news_finance_!_下周展望：双顶形成黄金面临大跌风险 美元技术面实现重大突破_!_趋势线,FXStreet,FX168,双顶,失业率,黄金
+6552229158854853128_!_104_!_news_finance_!_周末晒基金20180506_!_股市,月份第一,周测试
+6552341886009344525_!_104_!_news_finance_!_细数巴菲特股东大会干货，最不容错过的十大看点！_!_股东大会,Jain,Abel,伯克希尔哈撒韦,巴菲特,伯克希尔,中国股市
+6552365054509974019_!_104_!_news_finance_!_阿里18Q1：收入增长61%，成本上涨110%，均创历史新高_!_天猫,阿里巴巴,零售,盒马,阿里云
+6552397535128846862_!_104_!_news_finance_!_华锋股份：3个月内不再筹划除购买华创电动车股权之外重大事项_!_电动车,股票,股权,北京理工
+6552294683744141572_!_104_!_news_finance_!_什么药能同时在桃、李、柑橘、荔枝树上杀螨、蚧、夜蛾和卵？_!_
+6552382662953140743_!_104_!_news_finance_!_不只是水晶杯 捷克主宾国文化周带你领略不一样的捷克风情_!_义乌,捷克馆,文化周,主宾国,捷克
+6552288375703863822_!_104_!_news_finance_!_港元汇率前景仍然稳定 不同领域投资宜不同策略_!_联系汇率制,HIBOR,香港,港元,金管局,LIBOR
+6552456541851615501_!_106_!_news_house_!_广西北海的房价还有投资价值吗？_!_
+6552446831303852292_!_106_!_news_house_!_建房屋第一层最高可以多少米？_!_
+6552381502364057869_!_106_!_news_house_!_为什么房价低的时候没人买，房价猛涨大家却都抢着买？_!_
+6552375605596258567_!_106_!_news_house_!_未来房价会跌到什么程度？_!_
+6552467553061437699_!_106_!_news_house_!_为什么长春房价这么低？_!_
+6552143997052125444_!_106_!_news_house_!_中介卖房子如何找到客户？_!_
+6552380666770620676_!_106_!_news_house_!_已买房的人和未买房的人，可预期的3-5年内，生活会有什么不同？_!_
+6552180607286247940_!_106_!_news_house_!_房地产企业拿地热情降温 盈利空间持续受限_!_住宅用地,宅地,土地市场,城市土地,重点城市
+6552257668596105475_!_106_!_news_house_!_马云说十年后房价如葱，那我现在还能买房吗？_!_
+6552354088648966408_!_106_!_news_house_!_苏州房价您怎么看？_!_
+6552464255889703176_!_106_!_news_house_!_肇庆大旺高新区的龙光欢玖龙湖好还是保利爱乐小镇好？_!_
+6552428255226364164_!_106_!_news_house_!_买卖二手房，如何网签？_!_
+6552343403605000462_!_106_!_news_house_!_不考虑价格因素，你想住武汉的哪个小区？_!_
+6552423165828530440_!_106_!_news_house_!_乌鲁木齐周边哪有野钓？_!_
+6552394909075112199_!_106_!_news_house_!_2004年手里有100万，买腾讯股票和投资上海房子哪个赚的更多？_!_
+6552414100230504708_!_106_!_news_house_!_碧桂园精装房怎么样？_!_
+6552431001832784136_!_106_!_news_house_!_十年后房子真的降到普通人可以接受的二三十万一套吗？_!_
+6552435026187780365_!_106_!_news_house_!_武汉未来的房价会涨到100000元/平米吗？_!_
+6552490235870052877_!_106_!_news_house_!_王健林：最有钱的不是房地产，是银行！另一位听后笑而不语_!_王健林,万达,工商银行,银行,财富
+6552467399025623556_!_107_!_news_car_!_你的汽车钥匙还在“裸奔”吗？多功能双层钥匙包，霸气实用贼实用_!_汽车钥匙包,钥匙扣,钥匙包,韩版汽车钥匙包,多功能
+6552454075516977678_!_107_!_news_car_!_朋友35.22万提了辆奔驰GLC 还能有这价位？_!_R20,奔驰GLC,经销商,减速带,万元GLC
+6552263387609301512_!_107_!_news_car_!_这184位青海车主尽快接受违法处理，否则车辆将被强制报废_!_摩托车,西宁市,西宁,报废标准,中华人民共和国道路交通安全法
+6552300413285040647_!_107_!_news_car_!_车友：改装汽车排气管，会增加动力吗？_!_发动机,排气,高速运转时,排气管,发动机低速
+6552463830801187079_!_107_!_news_car_!_适合年轻人驾驶，动力性能较好，又保持在十五万以内的SUV有哪些？_!_
+6552317325125615885_!_107_!_news_car_!_宝马5好还是奥迪A6好？_!_
+6552298236269314318_!_107_!_news_car_!_炎热的夏天马上就要来了，如何给爱车提前做好保养？_!_
+6552192948409205262_!_107_!_news_car_!_夏天开车时，老司机都会带上这些降暑神器，实用又便携_!_遮阳罩,空调,车内,加湿器,车内环境
+6552341194368614916_!_107_!_news_car_!_「这倒霉劲儿！」沈阳轿车失控撞护栏 对向仨车全遭殃_!_沈阳,护栏
+6552271412042662148_!_107_!_news_car_!_奔驰GLC这车到底开起来怎么样，我只想听车主的真实用车感受？_!_
+6552354229820850701_!_107_!_news_car_!_锋芒新生 红旗H5深圳上市发布会_!_红旗,深圳锦龙红旗,红旗体验中心,中国一汽,红旗H5,一汽红旗,BLIS
+6552300547070755331_!_107_!_news_car_!_G7011十天高速徽天段成县站出口临时养护施工，请绕行！_!_
+6552342303069962760_!_107_!_news_car_!_迈锐宝XL，感受品质_!_BOSE,CarPlay,自适应,ADAS,扬声器
+6552127979521049096_!_107_!_news_car_!_中国品牌的骄傲，上汽大通打开英国市场！_!_EV80,MAXUS,LDV,上汽集团,上汽大通
+6552455029935047181_!_107_!_news_car_!_开车自驾游，同事建议带上这几样车品，坐享豪配之静美_!_便携式,洗车器,雪佛兰,mp3,HUD
+6552320510997299720_!_107_!_news_car_!_新一代车窗膜，功能太强悍！学会这一招，车内不开空调还清凉_!_空调出风口,车窗,户外活动,空调,汽车空调
+6552441672309932547_!_107_!_news_car_!_首辆长安换代CS35已上路，颜值力压宝骏510，不足7万合资车都急了_!_CS55,长安汽车,CS35,宝骏510,长安,长安CS35,宝骏,CS75
+6552346059908055566_!_107_!_news_car_!_一颗在骚动的心-爱丽舍，你值得开回家_!_信用卡,提车,0首付,新大陆,爱丽舍
+6552331275296833799_!_108_!_news_edu_!_如何有效的提高自己的精力，具体该做些什么如何去做呢？_!_
+6552365186236285191_!_108_!_news_edu_!_高考倒计时33天，考生如何迅速提分？_!_
+6552309430015230215_!_108_!_news_edu_!_为什么山东大学会逐渐没落？_!_
+6552307889015357959_!_108_!_news_edu_!_合肥三模文综，感觉考的怎么样啊？_!_
+6552279500648874504_!_108_!_news_edu_!_一篇文章教给你究竟该怎么背单词_!_背单词,合成词,万能公式,单词,衍生词
+6552356697115984388_!_108_!_news_edu_!_苍溪县城郊中学校举行广元三诊成绩分析暨质量表彰会_!_苍溪县,广元,四川省,教师
+6552394969619890440_!_108_!_news_edu_!_为什么状语从句变独立结构谓语要变成非谓语？_!_
+6552278300780134915_!_108_!_news_edu_!_谢谢您，马克思！_!_自然哲学,马克思主义,人类解放,哲学,马克思
+6552468298087268612_!_108_!_news_edu_!_孩子英语阅读理解不好，怎么办？_!_
+6552340583866696200_!_108_!_news_edu_!_江苏南京无锡常州博大五年制专转本捷报来袭：你还在傻傻的等待？_!_无锡,专转本2018江苏博大专转本2018江苏,江苏,年一贯制专转本,南京
+6552333001680749064_!_108_!_news_edu_!_张家界市一中举办庆“五四”青年节校园迷你马拉松活动_!_体育运动,马拉松,迷你马拉松,张家界市,张家界
+6552491697157177604_!_108_!_news_edu_!_如何成为一个好的思考者？_!_
+6552458746637845006_!_108_!_news_edu_!_北川中学，归来仍是少年_!_北川中学,北川
+6552277869534380558_!_108_!_news_edu_!_为进“民办”家长带娃冒雨“赶考” “面谈”题目曝光_!_神题,面谈,外国语,鲤鱼姐姐,民办
+6552288723327779080_!_108_!_news_edu_!_清华北大本科生或者一些211和985金融专业毕业的本科生年薪怎样？_!_
+6552360190224630276_!_108_!_news_edu_!_高性价比课程背后藏"阴阳合同“承诺退款却无道歉_!_业务员,新合同,性价比,培训课程,教育中心
+6552280332782010628_!_108_!_news_edu_!_清华北大本科生或者一些211和985金融专业毕业的本科生年薪怎么样？_!_
+6552299349387248131_!_108_!_news_edu_!_荥经县严道四小召开“法治进校园”专题会_!_专题会,荥经县,教师法,李宗平,中小学教师职业道德,中华人民共和国教育法
+6552411280357982472_!_109_!_news_tech_!_如何用ps很好的叠加色彩？_!_
+6552265668278878733_!_109_!_news_tech_!_第二届世界智能大会世界智能科技展邀你体验“黑科技”_!_机器人,光大集团,科大讯飞,智能家居,通用电气
+6552168160110313987_!_109_!_news_tech_!_聊聊shopify_!_网上商店,Shopify,跨境电商,独立站,托比亚斯
+6552464932175085831_!_109_!_news_tech_!_在台湾旅游，可以使用支付宝或者微信支付吗？_!_
+6552465348019355918_!_109_!_news_tech_!_一加6手机需要抢吗？一加的系统和小米比怎么样？_!_
+6552301724307030536_!_109_!_news_tech_!_这家银行够狠！半年内使用两次蚂蚁借呗京东白条 别想申房贷车贷_!_白条,杭州银行,征信,征信报告,个人消费贷款合同,银行
+6552292151802200590_!_109_!_news_tech_!_普宙无人机黄立：我们要用核心技术及安卓模式改变无人机市场格局_!_CES,黄立,安卓模式,闭门造车,智能手机,高德红外,大疆
+6552169326151991560_!_109_!_news_tech_!_农村电商，如何做好电商公共服务？_!_
+6552435333609292045_!_109_!_news_tech_!_贾跃亭新车回国，造车能救巨亏的乐视吗？_!_
+6552384456311702020_!_109_!_news_tech_!_关于ODE币经典技术成保障，可能成为币市新指向标的几点分析_!_虚拟货币市场,传统货币,比特币,区块链,ODE币
+6552458913411760388_!_109_!_news_tech_!_历史上有哪些恐怖的发明？_!_
+6552465314158739715_!_109_!_news_tech_!_微信支付是否已远远超过支付宝？_!_
+6552383485783310851_!_109_!_news_tech_!_互联网巨头品牌竞争拼“潮”劲，苏宁小狮子卖萌领潮流_!_吉祥物,小狮子,苏格拉宁,苏宁易购,苏宁
+6552431425843364355_!_109_!_news_tech_!_腾讯微视：一条自我衔尾的蛇_!_防御性产品,短视频,马化腾,腾讯微博,腾讯微视,内容,创造101,微视身上
+6552303231265931534_!_109_!_news_tech_!_相比苹果和国内的阿里腾讯，三星的强大体现在什么地方？_!_
+6552471964299034894_!_109_!_news_tech_!_华为荣耀7C和锤子3，选哪款比较好？_!_
+6552481207043490062_!_109_!_news_tech_!_3000到5000的手机，有什么好的推荐？_!_
+6552342659107651853_!_109_!_news_tech_!_你认为现有4G网，网速好不好？_!_
+6512670630276497677_!_110_!_news_military_!_历史上有哪些精准到可怕的预言？_!_德意志帝国,世界大战,沙皇俄国,奥匈帝国,恩格斯
+6537196620562301447_!_110_!_news_military_!_俄罗斯人用战斗机引擎组装的吹雪神器，没想到中国的更厉害！霸气_!_俄罗斯人,战斗机,俄罗斯,霸气
+6550187338377462276_!_110_!_news_military_!_2万吨核潜艇沉没，23名官兵苦撑8小时，最后一刻关闭核反应堆_!_苏俄,过氧化氢,库尔斯克号核潜艇,沉没,鱼雷
+6551960599046128132_!_110_!_news_military_!_它是山本特工队原型，造成我军一颗将星陨落，彭总下令全歼_!_全歼,李云龙,八路军,亮剑,益子挺进队,冈村宁次
+6501841775974416653_!_110_!_news_military_!_为什么没有国家向巴勒斯坦提供武器？_!_武器,美国,以色列,巴勒斯坦,军事实力
+6541198177238778372_!_110_!_news_military_!_洗澡应该尽量避开这四个时间，对身体非常不好，早知道早受益哦！_!_
+6552258559596298509_!_112_!_news_travel_!_苏州上方山森林动物世界鹦鹉笼舍为何闹鼠患？_!_
+6552307324613034247_!_112_!_news_travel_!_南京小吃街叫什么？_!_
+6552256835884155396_!_112_!_news_travel_!_湖北丹江口：3000人南水北调源头环库公路健身走_!_丹江口水库,丹江口库区,丹江口市,丹江口,三官殿
+6552282396530573572_!_112_!_news_travel_!_溪流与河流的差别是什么？_!_
+6552314414031175939_!_112_!_news_travel_!_国内哪一家航空公司的飞机餐最好吃？_!_
+6552456109221740808_!_112_!_news_travel_!_农家大门上可以用五脊六兽吗？_!_
+6552336954426393102_!_112_!_news_travel_!_这些鬼斧神工般的小村落，最后一个竟怪异到让人无法相信_!_大渡河,古路村,营盘乡,菊径村,三门峡
+6552313691335819789_!_112_!_news_travel_!_云南省5月6日11时至5月7日11时部分风景区天气预报_!_转阵雨,鸡足山,风景区,石林风景区,青龙峡风景区
+6552457495384687112_!_112_!_news_travel_!_一个人的朝圣（1）_!_哈罗德,说走就走,旅馆
+6552424675429843207_!_112_!_news_travel_!_《雍正王朝》四阿哥胤禛和十三阿哥胤祥，办差路过江夏镇为什么必须下马步行通过？_!_
+6552305458902729223_!_112_!_news_travel_!_想见识真正的老南通 到这个巷子来转一转_!_臭豆腐,寺街,南通市,南通,黄包车
+6552399429901484292_!_112_!_news_travel_!_广西北海和三亚哪个好玩？费用哪个高？_!_
+6552390933998993923_!_112_!_news_travel_!_这个国家最大的城市，还比不过中国一个镇_!_印度教,尼泊尔,加德满都,苦行僧,印度教徒
+6552335200402014734_!_112_!_news_travel_!_喜大普奔！历时3年，宁波又将添一地标性建筑！竣工时间在今年的_!_游泳馆,宁波,吴文奎,BIM,清水混凝土
+6552395692252332547_!_112_!_news_travel_!_泰安最美骑行路线，风景超美超好看，没骑过的赶紧去骑吧！_!_泰安,三大主题景观,天平湖,东湖,人间仙境
+6552486500028645639_!_112_!_news_travel_!_扶沟在郑州哪个地方？_!_
+6552467330671051271_!_112_!_news_travel_!_甘肃省社科联专家学者观摩临夏市旅游产业发展成果_!_习近平,古河州,民俗艺术,临夏市,八坊十三巷
+6552404579256893709_!_112_!_news_travel_!_被黑车宰是一种怎么样的体验？_!_
+6552474206909497869_!_112_!_news_travel_!_郴州飞天山+高椅岭，美得不一样的丹霞地貌，很值得一去_!_高椅岭,飞天山,丹霞地貌,地质公园,郴州
+6552367559658701325_!_112_!_news_travel_!_云南茶花甲天下，艳甲滇云在大理了_!_茶花,大理,大理古城,云南,甲滇云
+6552440761344852494_!_113_!_news_world_!_哥伦比亚排除地雷 清除和平障碍_!_哥伦比亚总统,爆炸物,哥伦比亚,桑托斯
+6552311509018476813_!_113_!_news_world_!_日本和以色列为什么会在贫瘠的土地上建立起高度发达的国家？_!_
+6552483261942071565_!_113_!_news_world_!_特朗普为什么可以随意废除此前美国政府达成的各种协议？_!_
+6552278526052008200_!_113_!_news_world_!_以现在海平面上涨速度，淹没日本需要多久，来中国是否收留？_!_
+6552347385824018692_!_113_!_news_world_!_如果特朗普遭弹劾下台，谁会来接替他当总统？_!_
+6552345528573624589_!_113_!_news_world_!_如果特朗普遭弹劾下台，谁会来接替他当总统？_!_
+6552277960408171011_!_113_!_news_world_!_叙利亚不再孤立无援，两大国家都在支援他，不用孤军奋战了！_!_伊朗,大马士革,叙利亚,普京,以色列
+6552228465804837123_!_113_!_news_world_!_哪位退休的教师领到百分之百退休金？_!_
+6552256940389433860_!_113_!_news_world_!_俄安全局：抓获5名“伊斯兰国”分支组织成员_!_近卫军,俄罗斯联邦安全局,Fontanka,雅罗斯拉夫尔,伊斯兰国
+6552324437411102984_!_113_!_news_world_!_为什么罗马帝国崩溃后，再也没有重新崛起，甚至长期陷入四分五裂？_!_
+6552361808668131598_!_113_!_news_world_!_美国建国不过两百年，为什么会成为世界强国？_!_
+6552453717327610371_!_113_!_news_world_!_关闭领空！重兵集结到位！美公开支持，以色列准备越境打击叙伊？_!_以色列,约旦河,叙利亚,伊朗,以色列总理
+6552176353930641672_!_113_!_news_world_!_为什么韩国人的身份证上有汉字？_!_
+6552438170758152461_!_113_!_news_world_!_美国建国不过两百年，为什么会成为世界强国？_!_
+6552268220517057031_!_113_!_news_world_!_英监管机构要求剑桥分析向美国公民交出数据_!_Carroll,美国公民,剑桥分析,SAR,数据保护法,ICO
+6552479163016544525_!_113_!_news_world_!_特朗普所做的一切会不会只是为了自己的商业帝国？_!_
+6552275276573704708_!_113_!_news_world_!_以色列已做好一切跨境攻打准备，正在寻求美国支持，叙伊危机来了_!_戈兰高地,叙利亚,以色列,伊朗,以色列总统
+6552447819645452804_!_115_!_news_agriculture_!_阳东大沟吉灶下村旧貌换新颜_!_吉灶下村,大沟镇,农家乐
+6552365566840013316_!_115_!_news_agriculture_!_61岁大爷挖了红薯精，小孩都抱不住，村民只有干瞪眼_!_简阳市,三轮车,红薯,四川省,烤红薯
+6551955594348069134_!_115_!_news_agriculture_!_农村俗语“会插不会插，瞅你两只脚”说的是什么意思？_!_
+6552452553634742798_!_115_!_news_agriculture_!_新疆沙漠脚下种稻子_!_库木塔格沙漠,旱稻,鄯善县,辟展乡,无人机
+6552361482867180035_!_115_!_news_agriculture_!_农村一种蔬菜，水陆都能长，北方吃梗，南方吃叶_!_北方人,蕹菜,南方人,空心菜,蔬菜
+6552345129208774915_!_115_!_news_agriculture_!_国家全面实施乡村振兴战略，面对农村人才流失有什么解决措施呢？_!_
+6552325745740349955_!_115_!_news_agriculture_!_一夜之间十佘亩西瓜全被开瓢，农户气得生病_!_种植户,西瓜,河北,大名县
+6552447947991155214_!_115_!_news_agriculture_!_2018.5.6｜生猪价格报_!_跌宁夏10.00华中河南省10.05,涨黑龙江9.85吉林省9.80,涨华南广西10.14,涨华北北京市10.03,涨全国均价10.18
+6552359120601612548_!_115_!_news_agriculture_!_定点养殖中蜂效益好吗？_!_
+6552127848914616835_!_115_!_news_agriculture_!_陕西特色小镇咋建 防止盲目造镇和地产化发展_!_行政建制镇,小镇,小城镇,意见,特色小镇,特色小城镇建设
+6552478264290443789_!_115_!_news_agriculture_!_农村小哥用这种方法捕鳝鱼龙虾，一个晚上收获56斤_!_营养价值,小龙虾,农村,鳝鱼,本草纲目,龙虾
+6552439935092130312_!_115_!_news_agriculture_!_仙居：文化墙扮靓美丽乡村_!_官路镇,文化墙,传统文化,仙居县,官路村
+6552460375147675907_!_115_!_news_agriculture_!_猪肉价格不断下跌，为什么农户还要坚持养猪？_!_
+6552408812408537608_!_115_!_news_agriculture_!_五月的乡村生长着一种似小番茄的果实，您可认识？有入药功效_!_
+6552391728727327239_!_115_!_news_agriculture_!_羊小反刍除了疫苗免疫，这些方法也管用！_!_药厂,小苏打,传染病,葡萄糖,反刍动物
+6552390501515919879_!_115_!_news_agriculture_!_国家将引进各类人才到农村创业，你准备了吗？谋划了吗？_!_农家乐,职称评定,职业农民,新型职业,乡村振兴
+6552323063352918542_!_115_!_news_agriculture_!_葡萄病虫害发生的绿色防控技术！要学会科学合理使用农药_!_炭疽病,可湿性粉剂,病斑,葡萄,霜霉病
+6551547654176768519_!_116_!_news_game_!_梦幻西游洗完野猪又洗蝙蝠 最后须弥炼妖居然有惊喜_!_梦幻西游
+6552428875488428551_!_116_!_news_game_!_DNF：那些充满回忆的地图，60版本地图大全！_!_格兰之森,悬空城,暗黑雷鸣废墟,烈焰格拉卡,地图
+6445082232641028365_!_116_!_news_game_!_主播小智长相很可怕吗？_!_脸吃饭,骂小智,小智,主播
+6552469298315526659_!_116_!_news_game_!_沉鱼落雁，闭月羞花，这4大美女在王者荣耀中应该是哪些呢？_!_王者荣耀,甄姬,大乔,王昭君,娜可露露
+6552299418568098051_!_116_!_news_game_!_Dnf武器怎么上13？_!_
+6552293748825391624_!_116_!_news_game_!_卷毛上分推荐：版本最恶心的辅助 索拉卡_!_E技能,艾黎,索拉卡
+6444028761334612237_!_116_!_news_game_!_王者荣耀为什么有人玩一个英雄过百局还是大坑？_!_安琪拉,机器人,英雄,王者荣耀,后羿
+6552397238302146819_!_116_!_news_game_!_刀塔2震中杯总决赛，你希望LGD vs冯提莫战队还是Liquid，为什么？_!_
+6552369554956222990_!_116_!_news_game_!_几乎每款动作游戏你都能见到这些技能，游戏技能抄袭问题司空见惯_!_火影忍者OL,游戏,最终幻想15,红丸OL,巫师3,上古世纪,逆战,龙爵,天劫,招式,唱吧炫舞,火球术,魔兽争霸：冰封王座,游戏中都
+6552495393173668110_!_116_!_news_game_!_LOL你见过哪些五个位置都能打的英雄？_!_
+6552394216650047751_!_116_!_news_game_!_如何评价首个以职业选手命名的比赛uzi杯？_!_
+6552331941465555207_!_116_!_news_game_!_《王者荣耀》会不会倒闭？_!_
+6552352146468110862_!_116_!_news_game_!_刺激战场玩家都喜欢玩的集中套路，一不小心就吃鸡了！_!_刺激战场,玩家
+6552472376334877187_!_116_!_news_game_!_Uzi本命VN惨遭黄金玩家各种单杀，诺手：我单杀Uzi，我可以吹了_!_德莱文,游戏,LPL,uzi,Uzi
+6552361451162436100_!_116_!_news_game_!_王者荣耀：在游戏中，要识时务者为俊杰，不要和这 4个英雄硬碰硬_!_一发子弹,游戏,孙尚香
+6528006878918607111_!_116_!_news_game_!_你为什么喜欢RNG？_!_
+6527241706197221640_!_116_!_news_game_!_一个游戏停服前是什么样子的，会出什么活动？_!_
+6552361667282338051_!_116_!_news_game_!_如果你喜欢《我的世界》，但是你妈妈又说《迷你世界》好，你会怎么办？_!_
+6552306091521212686_!_101_!_news_culture_!_杨过的人物形象是否受武松人物形象的影响？_!_
+6552336958943658247_!_101_!_news_culture_!_古代时刻表是怎样的，古时的午时三刻到底是11:45还是12：45？_!_
+6552295153342611981_!_101_!_news_culture_!_上联：一池绿水映明月；求下联_!_诗词欣赏,绿水,北固山
+6552338479265612039_!_101_!_news_culture_!_你最敬仰的伟人是谁？_!_
+6552318897943478532_!_101_!_news_culture_!_下联:人生苦短需何求？下联？_!_
+6552284896528695559_!_101_!_news_culture_!_青龙戏白虎，江涛乐夜风。如何接句？_!_
+6552484222932615438_!_101_!_news_culture_!_下联：两岸清风拂绿柳，你能对出上联吗？_!_
+6552250085466440206_!_101_!_news_culture_!_飘逸愈沉着，婀娜愈刚健——《书谱》用笔细节50例_!_书谱,用笔,书法,书法艺术
+6552269549335478541_!_101_!_news_culture_!_马克思是世界上最伟大的哲学家吗？_!_
+6552454585770836228_!_101_!_news_culture_!_上句：只身匹马孤剑陆游，下句如何对？_!_
+6552451573140685060_!_101_!_news_culture_!_上联：半夜风声惊旧梦，下联如何对？_!_
+6552479838739890445_!_101_!_news_culture_!_怎样临帖是最快最有效的？_!_
+6552460999650181636_!_101_!_news_culture_!_上善若水，这四个字到底什么意思？_!_本体论
+6552153617262969347_!_101_!_news_culture_!_《西游记》中的混世四猴，齐天大圣孙悟空属于哪一种？_!_通臂猿猴,六耳猕猴,孙悟空,赤尻马猴,灵明石猴,西游记
+6552407350177693960_!_101_!_news_culture_!_“全聚德”的“德”为什么少了一横？_!_
+6552320859346829571_!_101_!_news_culture_!_上联：蚕是天下虫，下联怎么对？_!_
+6552433755699544324_!_101_!_news_culture_!_太极拳的“双重”是指什么？为什么要避免双重？_!_
+6552448198126862606_!_101_!_news_culture_!_上联：上小下大尖山子，如何对下联？_!_
+6552383273123709192_!_101_!_news_culture_!_上联:苦菜花迎春花山菊花三花竞放，如何对下联？_!_
+6552372497453416968_!_102_!_news_entertainment_!_一亿粉丝的她，演戏不行，唱歌不行，凭啥成为中国粉丝最多的人？_!_王俊凯,快乐大本营,微博粉丝,谢娜,吉尼斯,粉丝
+6552431596903858695_!_102_!_news_entertainment_!_许玮甯近距离自拍 肩带滑落超性感_!_模特儿,社群网站,许玮甯,性感,电影圈
+6552364231302316552_!_102_!_news_entertainment_!_《温暖的弦》朱临路吃醋温柔被追称：原来这么多年，我爱的是你_!_朱临路,温暖的弦
+6552308030241767943_!_102_!_news_entertainment_!_当下炙手可热的童星，3岁便跟大腕搭戏，杨幂孙俪都夸她聪明可爱_!_虎妈猫爸,全民目击,仙剑奇侠传五,甄嬛传,小时代4,分手大师,万万没想到,杨幂,纪姿含,剧情片
+6552424296210235918_!_102_!_news_entertainment_!_霍启刚堂弟有了新恋情，对方竟是香港娱乐圈有名的“集邮女星”？_!_王秀琳,霍启文,香港娱乐圈,霍启刚,娱乐圈
+6552319625168683534_!_102_!_news_entertainment_!_李湘全家近照，90天甩肉30斤瘦成瓜子脸，女儿说英文被赞口音纯正_!_英语口语,王诗龄,王嘉尔,李湘,瓜子脸,爸爸去哪儿
+6552428387971891726_!_102_!_news_entertainment_!_《下一站，别离》秋阳是真讨厌，表白还要掉坑里，心疼盛夏_!_大结局,下一站，别离,真情流露,秋阳
+6552495692797968900_!_102_!_news_entertainment_!_体操世界冠军李小鹏一家四口近照曝光，孩子国籍却引发网友争议_!_李小鹏,奥利奥,李馨琪,体操,李小平
+6552438039912645133_!_102_!_news_entertainment_!_作为综艺嘉宾，他开创了娱乐圈的先例，让网友们都站到了统一战线_!_王艺诺,刘宪华,吐槽大会,李诞,何炅,张绍刚
+6552360542298702339_!_102_!_news_entertainment_!_赵本山老师出现抖音，一天涨粉200万，刘一手的评论亮了_!_小品,刘一手,春晚,赵本山老师,本山老师
+6552421494083813895_!_102_!_news_entertainment_!_付辛博颖儿大婚将至 网友：怎么觉得付辛博长得和高云翔越来越像_!_付辛博,高云翔越来越,颖儿大婚,付辛博颖儿大婚,付辛博颖儿
+6552366590833197582_!_102_!_news_entertainment_!_张翰和娜扎分手，郑爽胡彦斌说拜拜，都是被爱情伤害过的人！_!_全民目击,胡彦斌,郑爽,张翰,娜扎,相信爱情
+6552416737550139907_!_102_!_news_entertainment_!_成龙密谋见女儿吴卓林 却被这件事阻扰_!_吴绮莉,吴卓林,于成龙,Andi,成龙
+6552428369181409799_!_102_!_news_entertainment_!_郑爽的弹射1分钟ko两个对手，激动大叫，并拥抱队友，却被表白！_!_这！就是铁甲,张一山,撒贝宁,郑爽,综艺
+6552307894388261383_!_102_!_news_entertainment_!_徐濠萦现身街头，独一无二的造型，网友：不及陈冠希拽拽的一笑_!_独一无二,徐濠萦,不及陈冠希拽拽,陈冠希,笑徐濠萦现身街头
+6552452002939404814_!_102_!_news_entertainment_!_香港娱乐圈被“糟蹋”的5位女神：邱淑贞上榜，最后一位落下隐疾_!_Maggie Q,危情,刘銮雄,邱淑贞,关之琳,李嘉欣,赤裸特工
+6552409131003675144_!_102_!_news_entertainment_!_一个快被遗忘的男演员，恋爱10年隐婚4年了，如今4岁儿子长成这样_!_武工队传奇,蒋劲夫,谷智鑫,昭君出塞,我是特种兵,几度菊花香,恰同学少年,阿斯茹,兵出潼关,雷神突击
+6552356801445102093_!_102_!_news_entertainment_!_张翰微博高调表白张钧甯：有你的世界，一切才恢复色彩！_!_少年四大名捕,芒果台,张翰,伦敦眼,张钧甯,童话二分之一,偶像剧
+6552454171876917518_!_102_!_news_entertainment_!_《白鹿原》为什么没有火起来？_!_
+6552270624507560196_!_102_!_news_entertainment_!_《这就是街舞》第一季冠军最终被韩宇获得，你怎么看？_!_
+6552310112705315085_!_103_!_news_sports_!_骑士对猛龙G3，骑士詹姆斯8秒绝杀，骑士到了这个系列半决赛，你怎么看，骑士能夺冠吗？_!_
+6552350436873994509_!_103_!_news_sports_!_刘伟解说乒乓球比赛的水平如何？_!_
+6552452587545690627_!_103_!_news_sports_!_没有前场攻势如画的“恒大”和“上港”是我们想要的中超足球吗_!_足球,中超,恒大,转会费,孔卡
+6552404760689901827_!_103_!_news_sports_!_经常健身的人与不健身的人，整体上有何区别？_!_
+6552288985568248324_!_103_!_news_sports_!_雷霆三巨解散倒计时！湖人乔治最后1块碎片集满，甜瓜或加盟火箭_!_季后赛,湖人,乔治,雷霆,安东尼
+6552347828084015373_!_103_!_news_sports_!_如果刘国梁出国执教你是支持还是反对？_!_
+6552456169640689934_!_103_!_news_sports_!_骑士今年拿总冠军的概率有多大？_!_
+6552400121470910734_!_103_!_news_sports_!_金鱼能和鲤鱼混养吗？_!_
+6552459285702377731_!_103_!_news_sports_!_2019亚洲杯分组揭晓，国足抽中绝对好签，如何看待国足的分组？_!_
+6552402118924304643_!_103_!_news_sports_!_朱婷崴脚仍坎24分，瓦基弗3:2胜意甲晋级决赛，剑指冠军，你怎么看？_!_
+6552409455768633860_!_103_!_news_sports_!_四川或放弃哈达迪，球迷：来广东与阿联联手？_!_哈达迪,半决赛,广东队,阿联,广东
+6552488304799908103_!_103_!_news_sports_!_骑士对猛龙G3，詹姆斯投进压哨球，数万主场球迷沸腾JR激动到快要哭，猛龙球员眼神落寞，你怎么看？_!_
+6552436639547785735_!_103_!_news_sports_!_现役最自信的五大角色球员！追梦上榜，最后一位不虚任何人！_!_JR,超级巨星,斯蒂芬森,维特斯,詹姆斯
+6552327413387231491_!_103_!_news_sports_!_如何看待17-18赛季季后赛骑士VS猛龙系列赛中，骑士主教练泰伦卢的执教表现？_!_
+6552252197621465348_!_103_!_news_sports_!_钓鱼要什么技巧？_!_
+6552293215200870663_!_103_!_news_sports_!_山东鲁能有没有可能拿到今年的中超冠军？_!_
+6552337574457770503_!_103_!_news_sports_!_国乒暴露大问题！刘诗雯被2个主帅大赞，许昕放欧洲能打到40岁_!_许昕,刘诗雯,国乒,世锦赛,奥恰洛夫
+6552385630582604045_!_104_!_news_finance_!_上海的经济要想超越东京，需要多久？_!_
+6552275049015935235_!_104_!_news_finance_!_自3月23日以来，A股持续在3100点附近震荡，股市何时能突破该区域？_!_
+6552468374050308359_!_104_!_news_finance_!_准备职业炒股该从哪里入手？_!_
+6552366034060312846_!_104_!_news_finance_!_股票是怎么赚钱的？_!_
+6552407555593732616_!_104_!_news_finance_!_最近的股市，我们不妨可以学学这个西方的分析技术！_!_启明星,弹簧线,锤子线,支撑位,空方
+6552146563844538894_!_104_!_news_finance_!_一旦退休伯克希尔收购企业会变得困难？巴菲特：不担心_!_伯克希尔哈撒韦,巴菲特
+6552320439073374467_!_104_!_news_finance_!_山西大型国有企业是什么级别建制？_!_
+6552363494023365123_!_104_!_news_finance_!_芯片第一龙头横空出世，中报业绩暴增748倍，下周飙升成定局_!_集成电路,挠性电路,高技术,丹邦科技,柔性封装基板
+6552395858279661831_!_104_!_news_finance_!_想创业，投资在20万左右，一年利润在7万左右的行业有哪些？_!_
+6552339980448956676_!_104_!_news_finance_!_为什么证券市场风险较大，成功者廖廖，但是开户数不见减少？是因为门坎低的缘故吗？_!_
+6552437586181226766_!_104_!_news_finance_!_均线盘整阶段该如何处理？_!_
+6552229842455101956_!_104_!_news_finance_!_交通信用卡曲线提额技术揭秘，额度3万起步！_!_强制提,交通,信用卡
+6552370969514607117_!_104_!_news_finance_!_正新鸡排加盟费多少｜整店输出模式｜一站式服务_!_正新鸡排,食品店,风味小吃,鸡排,加盟费
diff --git a/cookbook/finetune_muti_classfication.ipynb b/cookbook/finetune_muti_classfication.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..531ded55ea0908578abde9ad8106ad981886387c
--- /dev/null
+++ b/cookbook/finetune_muti_classfication.ipynb
@@ -0,0 +1,2175 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "source": [
+    "# 使用微调的方法让模型对新闻分类更加准确\n",
+    "\n",
+    "在本实操手册中，开发者将使用ChatGLM3-6B base模型，对新闻分类数据集进行微调，并使用微调后的模型进行推理。\n",
+    "本操作手册使用公开数据集，数据集中包含了新闻标题和新闻关键词，开发者需要根据这些信息，将新闻分类到15个类别中的一个。\n",
+    "为了体现模型高效的学习能力，以及让用户更快的学习本手册，我们只使用了数据集中的一小部分数据，实际上，数据集中包含了超过40万条新闻数据。\n",
+    "\n",
+    "## 硬件要求\n",
+    "本实践手册需要使用 FP16 精度的模型进行推理，因此，我们推荐使用至少 16GB 显存的 英伟达 GPU 来完成本实践手册。\n"
+   ],
+   "metadata": {
+    "collapsed": false
+   }
+  },
+  {
+   "cell_type": "markdown",
+   "source": [
+    "首先，我们将原始的数据集格式转换为用于微调的`jsonl`格式，以方便进行微调。"
+   ],
+   "metadata": {
+    "collapsed": false
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "outputs": [],
+   "source": [
+    "import json\n",
+    "\n",
+    "# 路径可以根据实际情况修改\n",
+    "input_file_path = 'data/toutiao_cat_data_example.txt'\n",
+    "output_file_path = 'data/toutiao_cat_data_example.jsonl'\n",
+    "\n",
+    "# 提示词\n",
+    "prompt_prefix = \"\"\"\n",
+    "你是一个专业的新闻专家，请根据我提供的新闻信息，包括新闻标题，新闻关键词等信息，你要对每一行新闻类别进行分类并告诉我结果，不要返回其他信息和多于的文字，这些类别是:\n",
+    "news_story\n",
+    "news_culture\n",
+    "news_sports\n",
+    "news_finance\n",
+    "news_house\n",
+    "news_car\n",
+    "news_edu\n",
+    "news_tech\n",
+    "news_military\n",
+    "news_travel\n",
+    "news_world\n",
+    "stock\n",
+    "news_agriculture\n",
+    "news_game\n",
+    "请选择其中一个类别并返回，你只要返回类别的名称，不要返回其他信息。让我们开始吧:\n",
+    "\"\"\"\n",
+    "\n",
+    "# 分类代码和名称的映射\n",
+    "category_map = {\n",
+    "    \"100\": \"news_story\",\n",
+    "    \"101\": \"news_culture\",\n",
+    "    \"102\": \"news_entertainment\",\n",
+    "    \"103\": \"news_sports\",\n",
+    "    \"104\": \"news_finance\",\n",
+    "    \"106\": \"news_house\",\n",
+    "    \"107\": \"news_car\",\n",
+    "    \"108\": \"news_edu\",\n",
+    "    \"109\": \"news_tech\",\n",
+    "    \"110\": \"news_military\",\n",
+    "    \"112\": \"news_travel\",\n",
+    "    \"113\": \"news_world\",\n",
+    "    \"114\": \"stock\",\n",
+    "    \"115\": \"news_agriculture\",\n",
+    "    \"116\": \"news_game\"\n",
+    "}\n",
+    "\n",
+    "def process_line(line):\n",
+    "    # 分割每行数据\n",
+    "    parts = line.strip().split('_!_')\n",
+    "    if len(parts) != 5:\n",
+    "        return None\n",
+    "\n",
+    "    # 提取所需字段\n",
+    "    _, category_code, _, news_title, news_keywords = parts\n",
+    "\n",
+    "    # 构造 JSON 对象\n",
+    "    news_title = news_title if news_title else \"无\"\n",
+    "    news_keywords = news_keywords if news_keywords else \"无\"\n",
+    "    json_obj = {\n",
+    "        \"context\": prompt_prefix + f\"新闻标题: {news_title}\\n 新闻关键词: {news_keywords}\\n\",\n",
+    "        \"target\": category_map.get(category_code, \"无\")\n",
+    "    }\n",
+    "    return json_obj\n",
+    "\n",
+    "def convert_to_jsonl(input_path, output_path):\n",
+    "    with open(input_path, 'r', encoding='utf-8') as infile, \\\n",
+    "         open(output_path, 'w', encoding='utf-8') as outfile:\n",
+    "        for line in infile:\n",
+    "            json_obj = process_line(line)\n",
+    "            if json_obj:\n",
+    "                json_line = json.dumps(json_obj, ensure_ascii=False)\n",
+    "                outfile.write(json_line + '\\n')\n",
+    "\n",
+    "# 运行转换函数\n",
+    "convert_to_jsonl(input_file_path, output_file_path)"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-11-24T13:43:59.281779Z",
+     "end_time": "2023-11-24T13:43:59.330679Z"
+    }
+   }
+  },
+  {
+   "cell_type": "markdown",
+   "source": [
+    "## 使用没有微调的模型进行推理\n",
+    "首先，我们先试用原本的模基座模型进行推理，并查看效果。"
+   ],
+   "metadata": {
+    "collapsed": false
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "outputs": [],
+   "source": [
+    "PROMPT = \"\"\"\n",
+    "你是一个专业的新闻专家，请根据我提供的新闻信息，包括新闻标题，新闻关键词等信息，你要对每一行新闻类别进行分类并告诉我结果，不要返回其他信息和多于的文字，这些类别是:\n",
+    "news_story\n",
+    "news_culture\n",
+    "news_sports\n",
+    "news_finance\n",
+    "news_house\n",
+    "news_car\n",
+    "news_edu\n",
+    "news_tech\n",
+    "news_military\n",
+    "news_travel\n",
+    "news_world\n",
+    "stock\n",
+    "news_agriculture\n",
+    "news_game\n",
+    "请选择其中一个类别并返回，你只要返回类别的名称，不要返回其他信息。让我们开始吧:\n",
+    "新闻标题：华为手机扛下敌人子弹，是什么技术让其在战争中大放异彩？\n",
+    "新闻关键词: 华为手机\n",
+    "\"\"\""
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-11-24T13:43:59.377486Z",
+     "end_time": "2023-11-24T13:43:59.392276Z"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "Loading checkpoint shards:   0%|          | 0/7 [00:00<?, ?it/s]",
+      "application/vnd.jupyter.widget-view+json": {
+       "version_major": 2,
+       "version_minor": 0,
+       "model_id": "ed12c09cc7f443bab2f99b3ca7e99716"
+      }
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from transformers import AutoModel, AutoTokenizer\n",
+    "import  torch\n",
+    "# 参数设置\n",
+    "model_path = \"THUDM/chatglm3-6b-base\"\n",
+    "tokenizer_path = model_path\n",
+    "device = \"cuda\"\n",
+    "\n",
+    "tokenizer = AutoTokenizer.from_pretrained(tokenizer_path, trust_remote_code=True)\n",
+    "model = AutoModel.from_pretrained(model_path, load_in_8bit=False, trust_remote_code=True).to(device)"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-11-24T13:43:59.377486Z",
+     "end_time": "2023-11-24T13:44:11.180343Z"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "'news_house'"
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "max_new_tokens = 1024\n",
+    "temperature = 0.4\n",
+    "top_p = 0.9\n",
+    "inputs = tokenizer(PROMPT, return_tensors=\"pt\").to(device)\n",
+    "response = model.generate(input_ids=inputs[\"input_ids\"],max_new_tokens=max_new_tokens,temperature=temperature,top_p=top_p,do_sample=True)\n",
+    "response = response[0, inputs[\"input_ids\"].shape[-1]:]\n",
+    "origin_answer = tokenizer.decode(response, skip_special_tokens=True)\n",
+    "origin_answer"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-11-24T13:44:11.183956Z",
+     "end_time": "2023-11-24T13:44:12.313278Z"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "outputs": [],
+   "source": [
+    "del model\n",
+    "torch.cuda.empty_cache()"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-11-24T13:44:12.313278Z",
+     "end_time": "2023-11-24T13:44:12.463365Z"
+    }
+   }
+  },
+  {
+   "cell_type": "markdown",
+   "source": [
+    "我们可以发现，模型并没有正确的对新闻进行分类。这可能是由模型训练阶段的数据导致的问题。那么，我们通过微调这个模型，能不能实现更好的效果呢？"
+   ],
+   "metadata": {
+    "collapsed": false
+   }
+  },
+  {
+   "cell_type": "markdown",
+   "source": [
+    "## 使用官方的微调脚本进行微调\n",
+    "\n",
+    "在完成数据的切割之后，我们需要按照官方提供好的方式进行微调。我们使用的模型为`chatglm3-6b-base`基座模型，该模型相对于Chat模型，更容易上手微调，且更符合本章节的应用场景。\n",
+    "\n",
+    "我们将对应的参数设置好后，就可以直接执行下面的代码进行微调。该代码使用`Lora`方案进行微调，成本相较于全参微调大幅度降低。\n"
+   ],
+   "metadata": {
+    "collapsed": false
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "[2023-11-24 13:44:13,160] torch.distributed.run: [WARNING] master_addr is only used for static rdzv_backend and when rdzv_endpoint is not specified.\r\n",
+      "11/24/2023 13:44:15 - WARNING - __main__ - Process rank: 0, device: cuda:0, n_gpu: 1distributed training: True, 16-bits training: False\r\n",
+      "11/24/2023 13:44:15 - INFO - __main__ - Training/evaluation parameters Seq2SeqTrainingArguments(\r\n",
+      "_n_gpu=1,\r\n",
+      "adafactor=False,\r\n",
+      "adam_beta1=0.9,\r\n",
+      "adam_beta2=0.999,\r\n",
+      "adam_epsilon=1e-08,\r\n",
+      "auto_find_batch_size=False,\r\n",
+      "bf16=False,\r\n",
+      "bf16_full_eval=False,\r\n",
+      "data_seed=None,\r\n",
+      "dataloader_drop_last=False,\r\n",
+      "dataloader_num_workers=0,\r\n",
+      "dataloader_pin_memory=True,\r\n",
+      "ddp_backend=None,\r\n",
+      "ddp_broadcast_buffers=None,\r\n",
+      "ddp_bucket_cap_mb=None,\r\n",
+      "ddp_find_unused_parameters=None,\r\n",
+      "ddp_timeout=1800,\r\n",
+      "debug=[],\r\n",
+      "deepspeed=None,\r\n",
+      "disable_tqdm=False,\r\n",
+      "dispatch_batches=None,\r\n",
+      "do_eval=False,\r\n",
+      "do_predict=False,\r\n",
+      "do_train=False,\r\n",
+      "eval_accumulation_steps=None,\r\n",
+      "eval_delay=0,\r\n",
+      "eval_steps=None,\r\n",
+      "evaluation_strategy=no,\r\n",
+      "fp16=False,\r\n",
+      "fp16_backend=auto,\r\n",
+      "fp16_full_eval=False,\r\n",
+      "fp16_opt_level=O1,\r\n",
+      "fsdp=[],\r\n",
+      "fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_grad_ckpt': False},\r\n",
+      "fsdp_min_num_params=0,\r\n",
+      "fsdp_transformer_layer_cls_to_wrap=None,\r\n",
+      "full_determinism=False,\r\n",
+      "generation_config=None,\r\n",
+      "generation_max_length=None,\r\n",
+      "generation_num_beams=None,\r\n",
+      "gradient_accumulation_steps=2,\r\n",
+      "gradient_checkpointing=False,\r\n",
+      "gradient_checkpointing_kwargs=None,\r\n",
+      "greater_is_better=None,\r\n",
+      "group_by_length=False,\r\n",
+      "half_precision_backend=auto,\r\n",
+      "hub_always_push=False,\r\n",
+      "hub_model_id=None,\r\n",
+      "hub_private_repo=False,\r\n",
+      "hub_strategy=every_save,\r\n",
+      "hub_token=<HUB_TOKEN>,\r\n",
+      "ignore_data_skip=False,\r\n",
+      "include_inputs_for_metrics=False,\r\n",
+      "include_tokens_per_second=False,\r\n",
+      "jit_mode_eval=False,\r\n",
+      "label_names=None,\r\n",
+      "label_smoothing_factor=0.0,\r\n",
+      "learning_rate=2e-05,\r\n",
+      "length_column_name=length,\r\n",
+      "load_best_model_at_end=False,\r\n",
+      "local_rank=0,\r\n",
+      "log_level=passive,\r\n",
+      "log_level_replica=warning,\r\n",
+      "log_on_each_node=True,\r\n",
+      "logging_dir=output/news-20231124-134412-2e-05/runs/Nov24_13-44-15_iZwz90pbe3r8jeoaaobbf0Z,\r\n",
+      "logging_first_step=False,\r\n",
+      "logging_nan_inf_filter=True,\r\n",
+      "logging_steps=1.0,\r\n",
+      "logging_strategy=steps,\r\n",
+      "lr_scheduler_type=linear,\r\n",
+      "max_grad_norm=1.0,\r\n",
+      "max_steps=300,\r\n",
+      "metric_for_best_model=None,\r\n",
+      "mp_parameters=,\r\n",
+      "neftune_noise_alpha=None,\r\n",
+      "no_cuda=False,\r\n",
+      "num_train_epochs=3.0,\r\n",
+      "optim=adamw_torch,\r\n",
+      "optim_args=None,\r\n",
+      "output_dir=output/news-20231124-134412-2e-05,\r\n",
+      "overwrite_output_dir=False,\r\n",
+      "past_index=-1,\r\n",
+      "per_device_eval_batch_size=8,\r\n",
+      "per_device_train_batch_size=1,\r\n",
+      "predict_with_generate=False,\r\n",
+      "prediction_loss_only=False,\r\n",
+      "push_to_hub=False,\r\n",
+      "push_to_hub_model_id=None,\r\n",
+      "push_to_hub_organization=None,\r\n",
+      "push_to_hub_token=<PUSH_TO_HUB_TOKEN>,\r\n",
+      "ray_scope=last,\r\n",
+      "remove_unused_columns=True,\r\n",
+      "report_to=[],\r\n",
+      "resume_from_checkpoint=None,\r\n",
+      "run_name=output/news-20231124-134412-2e-05,\r\n",
+      "save_on_each_node=False,\r\n",
+      "save_safetensors=True,\r\n",
+      "save_steps=100,\r\n",
+      "save_strategy=steps,\r\n",
+      "save_total_limit=None,\r\n",
+      "seed=42,\r\n",
+      "skip_memory_metrics=True,\r\n",
+      "sortish_sampler=False,\r\n",
+      "split_batches=False,\r\n",
+      "tf32=None,\r\n",
+      "torch_compile=False,\r\n",
+      "torch_compile_backend=None,\r\n",
+      "torch_compile_mode=None,\r\n",
+      "torchdynamo=None,\r\n",
+      "tpu_metrics_debug=False,\r\n",
+      "tpu_num_cores=None,\r\n",
+      "use_cpu=False,\r\n",
+      "use_ipex=False,\r\n",
+      "use_legacy_prediction_loop=False,\r\n",
+      "use_mps_device=False,\r\n",
+      "warmup_ratio=0.0,\r\n",
+      "warmup_steps=0,\r\n",
+      "weight_decay=0.0,\r\n",
+      ")\r\n",
+      "[INFO|tokenization_utils_base.py:2020] 2023-11-24 13:44:15,353 >> loading file tokenizer.model\r\n",
+      "[INFO|tokenization_utils_base.py:2020] 2023-11-24 13:44:15,353 >> loading file added_tokens.json\r\n",
+      "[INFO|tokenization_utils_base.py:2020] 2023-11-24 13:44:15,353 >> loading file special_tokens_map.json\r\n",
+      "[INFO|tokenization_utils_base.py:2020] 2023-11-24 13:44:15,353 >> loading file tokenizer_config.json\r\n",
+      "[INFO|tokenization_utils_base.py:2020] 2023-11-24 13:44:15,353 >> loading file tokenizer.json\r\n",
+      "[INFO|configuration_utils.py:715] 2023-11-24 13:44:15,453 >> loading configuration file /Models/chatglm3-6b-base/config.json\r\n",
+      "[INFO|configuration_utils.py:715] 2023-11-24 13:44:15,454 >> loading configuration file /Models/chatglm3-6b-base/config.json\r\n",
+      "[INFO|configuration_utils.py:777] 2023-11-24 13:44:15,454 >> Model config ChatGLMConfig {\r\n",
+      "  \"_name_or_path\": \"/Models/chatglm3-6b-base\",\r\n",
+      "  \"add_bias_linear\": false,\r\n",
+      "  \"add_qkv_bias\": true,\r\n",
+      "  \"apply_query_key_layer_scaling\": true,\r\n",
+      "  \"apply_residual_connection_post_layernorm\": false,\r\n",
+      "  \"architectures\": [\r\n",
+      "    \"ChatGLMModel\"\r\n",
+      "  ],\r\n",
+      "  \"attention_dropout\": 0.0,\r\n",
+      "  \"attention_softmax_in_fp32\": true,\r\n",
+      "  \"auto_map\": {\r\n",
+      "    \"AutoConfig\": \"configuration_chatglm.ChatGLMConfig\",\r\n",
+      "    \"AutoModel\": \"modeling_chatglm.ChatGLMForConditionalGeneration\",\r\n",
+      "    \"AutoModelForCausalLM\": \"modeling_chatglm.ChatGLMForConditionalGeneration\",\r\n",
+      "    \"AutoModelForSeq2SeqLM\": \"modeling_chatglm.ChatGLMForConditionalGeneration\",\r\n",
+      "    \"AutoModelForSequenceClassification\": \"modeling_chatglm.ChatGLMForSequenceClassification\"\r\n",
+      "  },\r\n",
+      "  \"bias_dropout_fusion\": true,\r\n",
+      "  \"classifier_dropout\": null,\r\n",
+      "  \"eos_token_id\": 2,\r\n",
+      "  \"ffn_hidden_size\": 13696,\r\n",
+      "  \"fp32_residual_connection\": false,\r\n",
+      "  \"hidden_dropout\": 0.0,\r\n",
+      "  \"hidden_size\": 4096,\r\n",
+      "  \"kv_channels\": 128,\r\n",
+      "  \"layernorm_epsilon\": 1e-05,\r\n",
+      "  \"model_type\": \"chatglm\",\r\n",
+      "  \"multi_query_attention\": true,\r\n",
+      "  \"multi_query_group_num\": 2,\r\n",
+      "  \"num_attention_heads\": 32,\r\n",
+      "  \"num_layers\": 28,\r\n",
+      "  \"original_rope\": true,\r\n",
+      "  \"pad_token_id\": 0,\r\n",
+      "  \"padded_vocab_size\": 65024,\r\n",
+      "  \"post_layer_norm\": true,\r\n",
+      "  \"pre_seq_len\": null,\r\n",
+      "  \"prefix_projection\": false,\r\n",
+      "  \"quantization_bit\": 0,\r\n",
+      "  \"rmsnorm\": true,\r\n",
+      "  \"seq_length\": 32768,\r\n",
+      "  \"tie_word_embeddings\": false,\r\n",
+      "  \"torch_dtype\": \"float16\",\r\n",
+      "  \"transformers_version\": \"4.35.2\",\r\n",
+      "  \"use_cache\": true,\r\n",
+      "  \"vocab_size\": 65024\r\n",
+      "}\r\n",
+      "\r\n",
+      "[INFO|modeling_utils.py:3118] 2023-11-24 13:44:15,517 >> loading weights file /Models/chatglm3-6b-base/pytorch_model.bin.index.json\r\n",
+      "[INFO|configuration_utils.py:791] 2023-11-24 13:44:15,517 >> Generate config GenerationConfig {\r\n",
+      "  \"eos_token_id\": 2,\r\n",
+      "  \"pad_token_id\": 0\r\n",
+      "}\r\n",
+      "\r\n",
+      "Loading checkpoint shards: 100%|██████████████████| 7/7 [00:07<00:00,  1.02s/it]\r\n",
+      "[INFO|modeling_utils.py:3950] 2023-11-24 13:44:22,704 >> All model checkpoint weights were used when initializing ChatGLMForConditionalGeneration.\r\n",
+      "\r\n",
+      "[INFO|modeling_utils.py:3958] 2023-11-24 13:44:22,704 >> All the weights of ChatGLMForConditionalGeneration were initialized from the model checkpoint at /Models/chatglm3-6b-base.\r\n",
+      "If your task is similar to the task the model of the checkpoint was trained on, you can already use ChatGLMForConditionalGeneration for predictions without further training.\r\n",
+      "[INFO|modeling_utils.py:3525] 2023-11-24 13:44:22,706 >> Generation config file not found, using a generation config created from the model config.\r\n",
+      "Train dataset size: 4999\r\n",
+      "Sanity Check >>>>>>>>>>>>>\r\n",
+      "           '[gMASK]':  64790 ->   -100\r\n",
+      "               'sop':  64792 ->   -100\r\n",
+      "                  '':  30910 ->   -100\r\n",
+      "                '\\n':     13 ->   -100\r\n",
+      "                 '你':  54622 ->   -100\r\n",
+      "               '是一个':  32103 ->   -100\r\n",
+      "               '专业的':  34917 ->   -100\r\n",
+      "                '新闻':  31935 ->   -100\r\n",
+      "                '专家':  32114 ->   -100\r\n",
+      "                 '，':  31123 ->   -100\r\n",
+      "                 '请':  55073 ->   -100\r\n",
+      "                '根据':  31793 ->   -100\r\n",
+      "                 '我':  54546 ->   -100\r\n",
+      "               '提供的':  35100 ->   -100\r\n",
+      "                '新闻':  31935 ->   -100\r\n",
+      "                '信息':  31707 ->   -100\r\n",
+      "                 '，':  31123 ->   -100\r\n",
+      "                '包括':  31779 ->   -100\r\n",
+      "                '新闻':  31935 ->   -100\r\n",
+      "                '标题':  34490 ->   -100\r\n",
+      "                 '，':  31123 ->   -100\r\n",
+      "                '新闻':  31935 ->   -100\r\n",
+      "               '关键词':  35075 ->   -100\r\n",
+      "               '等信息':  46172 ->   -100\r\n",
+      "                 '，':  31123 ->   -100\r\n",
+      "                '你要':  34526 ->   -100\r\n",
+      "                 '对':  54570 ->   -100\r\n",
+      "                '每一':  32467 ->   -100\r\n",
+      "                 '行':  54560 ->   -100\r\n",
+      "                '新闻':  31935 ->   -100\r\n",
+      "                '类别':  38724 ->   -100\r\n",
+      "                '进行':  31636 ->   -100\r\n",
+      "                '分类':  33328 ->   -100\r\n",
+      "                 '并':  54724 ->   -100\r\n",
+      "               '告诉我':  38953 ->   -100\r\n",
+      "                '结果':  31951 ->   -100\r\n",
+      "                 '，':  31123 ->   -100\r\n",
+      "                '不要':  31844 ->   -100\r\n",
+      "                '返回':  34891 ->   -100\r\n",
+      "                '其他':  31722 ->   -100\r\n",
+      "               '信息和':  52701 ->   -100\r\n",
+      "                 '多':  54573 ->   -100\r\n",
+      "                 '于':  54579 ->   -100\r\n",
+      "               '的文字':  48746 ->   -100\r\n",
+      "                 '，':  31123 ->   -100\r\n",
+      "                '这些':  31704 ->   -100\r\n",
+      "                '类别':  38724 ->   -100\r\n",
+      "                 '是':  54532 ->   -100\r\n",
+      "                 ':':  30954 ->   -100\r\n",
+      "                '\\n':     13 ->   -100\r\n",
+      "              'news':   8480 ->   -100\r\n",
+      "                 '_':  30962 ->   -100\r\n",
+      "             'story':  12553 ->   -100\r\n",
+      "                '\\n':     13 ->   -100\r\n",
+      "              'news':   8480 ->   -100\r\n",
+      "                 '_':  30962 ->   -100\r\n",
+      "           'culture':  27458 ->   -100\r\n",
+      "                '\\n':     13 ->   -100\r\n",
+      "              'news':   8480 ->   -100\r\n",
+      "                 '_':  30962 ->   -100\r\n",
+      "                 's':  30917 ->   -100\r\n",
+      "             'ports':   3915 ->   -100\r\n",
+      "                '\\n':     13 ->   -100\r\n",
+      "              'news':   8480 ->   -100\r\n",
+      "                 '_':  30962 ->   -100\r\n",
+      "               'fin':   6242 ->   -100\r\n",
+      "              'ance':    562 ->   -100\r\n",
+      "                '\\n':     13 ->   -100\r\n",
+      "              'news':   8480 ->   -100\r\n",
+      "                 '_':  30962 ->   -100\r\n",
+      "             'house':   4199 ->   -100\r\n",
+      "                '\\n':     13 ->   -100\r\n",
+      "              'news':   8480 ->   -100\r\n",
+      "                 '_':  30962 ->   -100\r\n",
+      "               'car':   6747 ->   -100\r\n",
+      "                '\\n':     13 ->   -100\r\n",
+      "              'news':   8480 ->   -100\r\n",
+      "                 '_':  30962 ->   -100\r\n",
+      "               'edu':   7473 ->   -100\r\n",
+      "                '\\n':     13 ->   -100\r\n",
+      "              'news':   8480 ->   -100\r\n",
+      "                 '_':  30962 ->   -100\r\n",
+      "              'tech':  12232 ->   -100\r\n",
+      "                '\\n':     13 ->   -100\r\n",
+      "              'news':   8480 ->   -100\r\n",
+      "                 '_':  30962 ->   -100\r\n",
+      "               'mil':  20477 ->   -100\r\n",
+      "             'itary':   2733 ->   -100\r\n",
+      "                '\\n':     13 ->   -100\r\n",
+      "              'news':   8480 ->   -100\r\n",
+      "                 '_':  30962 ->   -100\r\n",
+      "               'tra':   7441 ->   -100\r\n",
+      "               'vel':    609 ->   -100\r\n",
+      "                '\\n':     13 ->   -100\r\n",
+      "              'news':   8480 ->   -100\r\n",
+      "                 '_':  30962 ->   -100\r\n",
+      "             'world':   8515 ->   -100\r\n",
+      "                '\\n':     13 ->   -100\r\n",
+      "             'stock':  14148 ->   -100\r\n",
+      "                '\\n':     13 ->   -100\r\n",
+      "              'news':   8480 ->   -100\r\n",
+      "                 '_':  30962 ->   -100\r\n",
+      "                'ag':    369 ->   -100\r\n",
+      "               'ric':    995 ->   -100\r\n",
+      "            'ulture':   4768 ->   -100\r\n",
+      "                '\\n':     13 ->   -100\r\n",
+      "              'news':   8480 ->   -100\r\n",
+      "                 '_':  30962 ->   -100\r\n",
+      "              'game':   8947 ->   -100\r\n",
+      "                '\\n':     13 ->   -100\r\n",
+      "                 '请':  55073 ->   -100\r\n",
+      "                '选择':  31768 ->   -100\r\n",
+      "              '其中一个':  46753 ->   -100\r\n",
+      "                '类别':  38724 ->   -100\r\n",
+      "                 '并':  54724 ->   -100\r\n",
+      "                '返回':  34891 ->   -100\r\n",
+      "                 '，':  31123 ->   -100\r\n",
+      "                 '你':  54622 ->   -100\r\n",
+      "                '只要':  32100 ->   -100\r\n",
+      "                '返回':  34891 ->   -100\r\n",
+      "                 '类':  54931 ->   -100\r\n",
+      "                '别的':  34752 ->   -100\r\n",
+      "                '名称':  33624 ->   -100\r\n",
+      "                 '，':  31123 ->   -100\r\n",
+      "                '不要':  31844 ->   -100\r\n",
+      "                '返回':  34891 ->   -100\r\n",
+      "                '其他':  31722 ->   -100\r\n",
+      "                '信息':  31707 ->   -100\r\n",
+      "                 '。':  31155 ->   -100\r\n",
+      "               '让我们':  32817 ->   -100\r\n",
+      "                '开始':  31699 ->   -100\r\n",
+      "                 '吧':  55370 ->   -100\r\n",
+      "                 ':':  30954 ->   -100\r\n",
+      "                '\\n':     13 ->   -100\r\n",
+      "                '新闻':  31935 ->   -100\r\n",
+      "                '标题':  34490 ->   -100\r\n",
+      "                 ':':  30954 ->   -100\r\n",
+      "                  '':  30910 ->   -100\r\n",
+      "                '京城':  46921 ->   -100\r\n",
+      "                 '最':  54628 ->   -100\r\n",
+      "                '值得':  32421 ->   -100\r\n",
+      "                '你来':  52586 ->   -100\r\n",
+      "                 '场':  54686 ->   -100\r\n",
+      "                '文化':  31653 ->   -100\r\n",
+      "                '之旅':  35383 ->   -100\r\n",
+      "                 '的':  54530 ->   -100\r\n",
+      "               '博物馆':  32964 ->   -100\r\n",
+      "                '\\n':     13 ->   -100\r\n",
+      "                '新闻':  45302 ->   -100\r\n",
+      "               '关键词':  35075 ->   -100\r\n",
+      "                 ':':  30954 ->   -100\r\n",
+      "                  '':  30910 ->   -100\r\n",
+      "                '保利':  46340 ->   -100\r\n",
+      "                '集团':  31839 ->   -100\r\n",
+      "                 ',':  30932 ->   -100\r\n",
+      "                 '马':  54988 ->   -100\r\n",
+      "                 '未':  54933 ->   -100\r\n",
+      "                 '都':  54606 ->   -100\r\n",
+      "                 ',':  30932 ->   -100\r\n",
+      "                '中国':  31626 ->   -100\r\n",
+      "              '科学技术':  35587 ->   -100\r\n",
+      "                 '馆':  55294 ->   -100\r\n",
+      "                 ',':  30932 ->   -100\r\n",
+      "               '博物馆':  32964 ->   -100\r\n",
+      "                 ',':  30932 ->   -100\r\n",
+      "               '新中国':  35873 ->   -100\r\n",
+      "                '\\n':     13 ->   -100\r\n",
+      "              'news':   2374 ->   2374\r\n",
+      "                 '_':  30962 ->  30962\r\n",
+      "           'culture':  27458 ->  27458\r\n",
+      "                  '':      2 ->      2\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "                  '':      0 ->   -100\r\n",
+      "<<<<<<<<<<<<< Sanity Check\r\n",
+      "[INFO|trainer.py:544] 2023-11-24 13:44:24,838 >> max_steps is given, it will override any value given in num_train_epochs\r\n",
+      "[INFO|trainer.py:1723] 2023-11-24 13:44:25,787 >> ***** Running training *****\r\n",
+      "[INFO|trainer.py:1724] 2023-11-24 13:44:25,787 >>   Num examples = 4,999\r\n",
+      "[INFO|trainer.py:1725] 2023-11-24 13:44:25,787 >>   Num Epochs = 1\r\n",
+      "[INFO|trainer.py:1726] 2023-11-24 13:44:25,787 >>   Instantaneous batch size per device = 1\r\n",
+      "[INFO|trainer.py:1729] 2023-11-24 13:44:25,787 >>   Total train batch size (w. parallel, distributed & accumulation) = 2\r\n",
+      "[INFO|trainer.py:1730] 2023-11-24 13:44:25,787 >>   Gradient Accumulation steps = 2\r\n",
+      "[INFO|trainer.py:1731] 2023-11-24 13:44:25,788 >>   Total optimization steps = 300\r\n",
+      "[INFO|trainer.py:1732] 2023-11-24 13:44:25,788 >>   Number of trainable parameters = 1,949,696\r\n",
+      "  0%|                                                   | 0/300 [00:00<?, ?it/s][W reducer.cpp:1346] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters in the forward pass. This flag results in an extra traversal of the autograd graph every iteration,  which can adversely affect performance. If your model indeed never has any unused parameters in the forward pass, consider turning this flag off. Note that this warning may be a false positive if your model has flow control causing later iterations to have unused parameters. (function operator())\r\n",
+      "{'loss': 0.6584, 'learning_rate': 1.9933333333333334e-05, 'epoch': 0.0}         \r\n",
+      "{'loss': 1.1299, 'learning_rate': 1.9866666666666667e-05, 'epoch': 0.0}         \r\n",
+      "{'loss': 1.2722, 'learning_rate': 1.98e-05, 'epoch': 0.0}                       \r\n",
+      "{'loss': 0.5076, 'learning_rate': 1.9733333333333336e-05, 'epoch': 0.0}         \r\n",
+      "{'loss': 1.5129, 'learning_rate': 1.9666666666666666e-05, 'epoch': 0.0}         \r\n",
+      "{'loss': 1.7524, 'learning_rate': 1.9600000000000002e-05, 'epoch': 0.0}         \r\n",
+      "{'loss': 0.7554, 'learning_rate': 1.9533333333333335e-05, 'epoch': 0.0}         \r\n",
+      "{'loss': 0.834, 'learning_rate': 1.9466666666666668e-05, 'epoch': 0.0}          \r\n",
+      "{'loss': 2.0967, 'learning_rate': 1.94e-05, 'epoch': 0.0}                       \r\n",
+      "{'loss': 0.2806, 'learning_rate': 1.9333333333333333e-05, 'epoch': 0.0}         \r\n",
+      "{'loss': 1.0696, 'learning_rate': 1.926666666666667e-05, 'epoch': 0.0}          \r\n",
+      "{'loss': 0.5934, 'learning_rate': 1.9200000000000003e-05, 'epoch': 0.0}         \r\n",
+      "{'loss': 0.8784, 'learning_rate': 1.9133333333333335e-05, 'epoch': 0.01}        \r\n",
+      "{'loss': 1.9795, 'learning_rate': 1.9066666666666668e-05, 'epoch': 0.01}        \r\n",
+      "{'loss': 0.953, 'learning_rate': 1.9e-05, 'epoch': 0.01}                        \r\n",
+      "{'loss': 0.1516, 'learning_rate': 1.8933333333333334e-05, 'epoch': 0.01}        \r\n",
+      "{'loss': 0.7275, 'learning_rate': 1.886666666666667e-05, 'epoch': 0.01}         \r\n",
+      "{'loss': 0.2672, 'learning_rate': 1.88e-05, 'epoch': 0.01}                      \r\n",
+      "{'loss': 0.152, 'learning_rate': 1.8733333333333336e-05, 'epoch': 0.01}         \r\n",
+      "{'loss': 0.3321, 'learning_rate': 1.866666666666667e-05, 'epoch': 0.01}         \r\n",
+      "{'loss': 0.5358, 'learning_rate': 1.86e-05, 'epoch': 0.01}                      \r\n",
+      "{'loss': 0.3262, 'learning_rate': 1.8533333333333334e-05, 'epoch': 0.01}        \r\n",
+      "{'loss': 0.3477, 'learning_rate': 1.8466666666666667e-05, 'epoch': 0.01}        \r\n",
+      "{'loss': 0.2667, 'learning_rate': 1.8400000000000003e-05, 'epoch': 0.01}        \r\n",
+      "{'loss': 0.3765, 'learning_rate': 1.8333333333333333e-05, 'epoch': 0.01}        \r\n",
+      "{'loss': 0.1945, 'learning_rate': 1.826666666666667e-05, 'epoch': 0.01}         \r\n",
+      "{'loss': 0.2321, 'learning_rate': 1.8200000000000002e-05, 'epoch': 0.01}        \r\n",
+      "{'loss': 0.094, 'learning_rate': 1.8133333333333335e-05, 'epoch': 0.01}         \r\n",
+      "{'loss': 0.2454, 'learning_rate': 1.8066666666666668e-05, 'epoch': 0.01}        \r\n",
+      "{'loss': 0.0523, 'learning_rate': 1.8e-05, 'epoch': 0.01}                       \r\n",
+      "{'loss': 0.9747, 'learning_rate': 1.7933333333333333e-05, 'epoch': 0.01}        \r\n",
+      "{'loss': 0.7521, 'learning_rate': 1.7866666666666666e-05, 'epoch': 0.01}        \r\n",
+      "{'loss': 0.7373, 'learning_rate': 1.7800000000000002e-05, 'epoch': 0.01}        \r\n",
+      "{'loss': 0.2056, 'learning_rate': 1.7733333333333335e-05, 'epoch': 0.01}        \r\n",
+      "{'loss': 0.0107, 'learning_rate': 1.7666666666666668e-05, 'epoch': 0.01}        \r\n",
+      "{'loss': 0.0525, 'learning_rate': 1.76e-05, 'epoch': 0.01}                      \r\n",
+      "{'loss': 0.0377, 'learning_rate': 1.7533333333333337e-05, 'epoch': 0.01}        \r\n",
+      "{'loss': 0.8701, 'learning_rate': 1.7466666666666667e-05, 'epoch': 0.02}        \r\n",
+      "{'loss': 0.1365, 'learning_rate': 1.7400000000000003e-05, 'epoch': 0.02}        \r\n",
+      "{'loss': 0.763, 'learning_rate': 1.7333333333333336e-05, 'epoch': 0.02}         \r\n",
+      "{'loss': 0.6814, 'learning_rate': 1.726666666666667e-05, 'epoch': 0.02}         \r\n",
+      "{'loss': 0.066, 'learning_rate': 1.72e-05, 'epoch': 0.02}                       \r\n",
+      "{'loss': 0.543, 'learning_rate': 1.7133333333333334e-05, 'epoch': 0.02}         \r\n",
+      "{'loss': 0.1186, 'learning_rate': 1.706666666666667e-05, 'epoch': 0.02}         \r\n",
+      "{'loss': 0.0753, 'learning_rate': 1.7e-05, 'epoch': 0.02}                       \r\n",
+      "{'loss': 0.1376, 'learning_rate': 1.6933333333333336e-05, 'epoch': 0.02}        \r\n",
+      "{'loss': 0.0322, 'learning_rate': 1.686666666666667e-05, 'epoch': 0.02}         \r\n",
+      "{'loss': 0.2514, 'learning_rate': 1.6800000000000002e-05, 'epoch': 0.02}        \r\n",
+      "{'loss': 0.9785, 'learning_rate': 1.6733333333333335e-05, 'epoch': 0.02}        \r\n",
+      "{'loss': 0.4355, 'learning_rate': 1.6666666666666667e-05, 'epoch': 0.02}        \r\n",
+      "{'loss': 0.064, 'learning_rate': 1.66e-05, 'epoch': 0.02}                       \r\n",
+      "{'loss': 0.0915, 'learning_rate': 1.6533333333333333e-05, 'epoch': 0.02}        \r\n",
+      "{'loss': 0.0999, 'learning_rate': 1.646666666666667e-05, 'epoch': 0.02}         \r\n",
+      "{'loss': 0.0058, 'learning_rate': 1.64e-05, 'epoch': 0.02}                      \r\n",
+      "{'loss': 0.3339, 'learning_rate': 1.6333333333333335e-05, 'epoch': 0.02}        \r\n",
+      "{'loss': 0.604, 'learning_rate': 1.6266666666666668e-05, 'epoch': 0.02}         \r\n",
+      "{'loss': 0.3101, 'learning_rate': 1.62e-05, 'epoch': 0.02}                      \r\n",
+      "{'loss': 0.149, 'learning_rate': 1.6133333333333334e-05, 'epoch': 0.02}         \r\n",
+      "{'loss': 0.1271, 'learning_rate': 1.606666666666667e-05, 'epoch': 0.02}         \r\n",
+      "{'loss': 0.2854, 'learning_rate': 1.6000000000000003e-05, 'epoch': 0.02}        \r\n",
+      "{'loss': 0.3322, 'learning_rate': 1.5933333333333336e-05, 'epoch': 0.02}        \r\n",
+      "{'loss': 0.0556, 'learning_rate': 1.586666666666667e-05, 'epoch': 0.02}         \r\n",
+      "{'loss': 0.1277, 'learning_rate': 1.58e-05, 'epoch': 0.03}                      \r\n",
+      "{'loss': 1.4146, 'learning_rate': 1.5733333333333334e-05, 'epoch': 0.03}        \r\n",
+      "{'loss': 0.0079, 'learning_rate': 1.5666666666666667e-05, 'epoch': 0.03}        \r\n",
+      "{'loss': 0.0576, 'learning_rate': 1.5600000000000003e-05, 'epoch': 0.03}        \r\n",
+      "{'loss': 0.049, 'learning_rate': 1.5533333333333333e-05, 'epoch': 0.03}         \r\n",
+      "{'loss': 0.4167, 'learning_rate': 1.546666666666667e-05, 'epoch': 0.03}         \r\n",
+      "{'loss': 0.1903, 'learning_rate': 1.54e-05, 'epoch': 0.03}                      \r\n",
+      "{'loss': 0.0972, 'learning_rate': 1.5333333333333334e-05, 'epoch': 0.03}        \r\n",
+      "{'loss': 0.056, 'learning_rate': 1.5266666666666667e-05, 'epoch': 0.03}         \r\n",
+      "{'loss': 0.0937, 'learning_rate': 1.5200000000000002e-05, 'epoch': 0.03}        \r\n",
+      "{'loss': 0.0168, 'learning_rate': 1.5133333333333335e-05, 'epoch': 0.03}        \r\n",
+      "{'loss': 0.4977, 'learning_rate': 1.5066666666666668e-05, 'epoch': 0.03}        \r\n",
+      "{'loss': 0.7477, 'learning_rate': 1.5000000000000002e-05, 'epoch': 0.03}        \r\n",
+      "{'loss': 0.0054, 'learning_rate': 1.4933333333333335e-05, 'epoch': 0.03}        \r\n",
+      "{'loss': 0.0664, 'learning_rate': 1.4866666666666668e-05, 'epoch': 0.03}        \r\n",
+      "{'loss': 0.1106, 'learning_rate': 1.48e-05, 'epoch': 0.03}                      \r\n",
+      "{'loss': 0.3886, 'learning_rate': 1.4733333333333335e-05, 'epoch': 0.03}        \r\n",
+      "{'loss': 0.0432, 'learning_rate': 1.4666666666666666e-05, 'epoch': 0.03}        \r\n",
+      "{'loss': 0.9208, 'learning_rate': 1.46e-05, 'epoch': 0.03}                      \r\n",
+      "{'loss': 0.047, 'learning_rate': 1.4533333333333335e-05, 'epoch': 0.03}         \r\n",
+      "{'loss': 0.184, 'learning_rate': 1.4466666666666668e-05, 'epoch': 0.03}         \r\n",
+      "{'loss': 0.0166, 'learning_rate': 1.4400000000000001e-05, 'epoch': 0.03}        \r\n",
+      "{'loss': 0.0051, 'learning_rate': 1.4333333333333334e-05, 'epoch': 0.03}        \r\n",
+      "{'loss': 0.0221, 'learning_rate': 1.4266666666666668e-05, 'epoch': 0.03}        \r\n",
+      "{'loss': 0.0343, 'learning_rate': 1.4200000000000001e-05, 'epoch': 0.03}        \r\n",
+      "{'loss': 0.0725, 'learning_rate': 1.4133333333333334e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.612, 'learning_rate': 1.4066666666666669e-05, 'epoch': 0.04}         \r\n",
+      "{'loss': 0.0592, 'learning_rate': 1.4e-05, 'epoch': 0.04}                       \r\n",
+      "{'loss': 0.3324, 'learning_rate': 1.3933333333333334e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.4796, 'learning_rate': 1.3866666666666669e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.0342, 'learning_rate': 1.38e-05, 'epoch': 0.04}                      \r\n",
+      "{'loss': 0.1784, 'learning_rate': 1.3733333333333335e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.007, 'learning_rate': 1.3666666666666667e-05, 'epoch': 0.04}         \r\n",
+      "{'loss': 0.1859, 'learning_rate': 1.3600000000000002e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.1524, 'learning_rate': 1.3533333333333333e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.0107, 'learning_rate': 1.3466666666666668e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.2768, 'learning_rate': 1.3400000000000002e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.0235, 'learning_rate': 1.3333333333333333e-05, 'epoch': 0.04}        \r\n",
+      " 33%|█████████████▋                           | 100/300 [02:29<04:59,  1.50s/it][INFO|tokenization_utils_base.py:2428] 2023-11-24 13:46:55,413 >> tokenizer config file saved in output/news-20231124-134412-2e-05/checkpoint-100/tokenizer_config.json\r\n",
+      "[INFO|tokenization_utils_base.py:2437] 2023-11-24 13:46:55,413 >> Special tokens file saved in output/news-20231124-134412-2e-05/checkpoint-100/special_tokens_map.json\r\n",
+      "{'loss': 0.0081, 'learning_rate': 1.3266666666666668e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.0147, 'learning_rate': 1.3200000000000002e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.0156, 'learning_rate': 1.3133333333333334e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.0188, 'learning_rate': 1.3066666666666668e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.1504, 'learning_rate': 1.3000000000000001e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.1548, 'learning_rate': 1.2933333333333334e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.1723, 'learning_rate': 1.2866666666666667e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.0258, 'learning_rate': 1.2800000000000001e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.0329, 'learning_rate': 1.2733333333333336e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.0042, 'learning_rate': 1.2666666666666667e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.2594, 'learning_rate': 1.2600000000000001e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.3348, 'learning_rate': 1.2533333333333336e-05, 'epoch': 0.04}        \r\n",
+      "{'loss': 0.1135, 'learning_rate': 1.2466666666666667e-05, 'epoch': 0.05}        \r\n",
+      "{'loss': 0.0225, 'learning_rate': 1.2400000000000002e-05, 'epoch': 0.05}        \r\n",
+      "{'loss': 0.0036, 'learning_rate': 1.2333333333333334e-05, 'epoch': 0.05}        \r\n",
+      "{'loss': 0.0105, 'learning_rate': 1.2266666666666667e-05, 'epoch': 0.05}        \r\n",
+      "{'loss': 0.0565, 'learning_rate': 1.22e-05, 'epoch': 0.05}                      \r\n",
+      "{'loss': 0.0286, 'learning_rate': 1.2133333333333335e-05, 'epoch': 0.05}        \r\n",
+      "{'loss': 0.5644, 'learning_rate': 1.206666666666667e-05, 'epoch': 0.05}         \r\n",
+      "{'loss': 0.3605, 'learning_rate': 1.2e-05, 'epoch': 0.05}                       \r\n",
+      "{'loss': 0.6441, 'learning_rate': 1.1933333333333335e-05, 'epoch': 0.05}        \r\n",
+      "{'loss': 0.0111, 'learning_rate': 1.186666666666667e-05, 'epoch': 0.05}         \r\n",
+      "{'loss': 0.0098, 'learning_rate': 1.18e-05, 'epoch': 0.05}                      \r\n",
+      "{'loss': 0.9559, 'learning_rate': 1.1733333333333335e-05, 'epoch': 0.05}        \r\n",
+      "{'loss': 0.0042, 'learning_rate': 1.1666666666666668e-05, 'epoch': 0.05}        \r\n",
+      "{'loss': 0.4136, 'learning_rate': 1.16e-05, 'epoch': 0.05}                      \r\n",
+      "{'loss': 0.5268, 'learning_rate': 1.1533333333333334e-05, 'epoch': 0.05}        \r\n",
+      "{'loss': 0.1325, 'learning_rate': 1.1466666666666668e-05, 'epoch': 0.05}        \r\n",
+      "{'loss': 0.0051, 'learning_rate': 1.14e-05, 'epoch': 0.05}                      \r\n",
+      "{'loss': 0.0006, 'learning_rate': 1.1333333333333334e-05, 'epoch': 0.05}        \r\n",
+      "{'loss': 0.007, 'learning_rate': 1.1266666666666668e-05, 'epoch': 0.05}         \r\n",
+      "{'loss': 0.441, 'learning_rate': 1.1200000000000001e-05, 'epoch': 0.05}         \r\n",
+      "{'loss': 0.0102, 'learning_rate': 1.1133333333333334e-05, 'epoch': 0.05}        \r\n",
+      "{'loss': 0.0415, 'learning_rate': 1.1066666666666669e-05, 'epoch': 0.05}        \r\n",
+      "{'loss': 0.0206, 'learning_rate': 1.1000000000000001e-05, 'epoch': 0.05}        \r\n",
+      "{'loss': 0.1697, 'learning_rate': 1.0933333333333334e-05, 'epoch': 0.05}        \r\n",
+      "{'loss': 0.2054, 'learning_rate': 1.0866666666666667e-05, 'epoch': 0.05}        \r\n",
+      "{'loss': 0.005, 'learning_rate': 1.0800000000000002e-05, 'epoch': 0.06}         \r\n",
+      "{'loss': 0.9604, 'learning_rate': 1.0733333333333333e-05, 'epoch': 0.06}        \r\n",
+      "{'loss': 0.0043, 'learning_rate': 1.0666666666666667e-05, 'epoch': 0.06}        \r\n",
+      "{'loss': 0.4815, 'learning_rate': 1.0600000000000002e-05, 'epoch': 0.06}        \r\n",
+      "{'loss': 0.4817, 'learning_rate': 1.0533333333333333e-05, 'epoch': 0.06}        \r\n",
+      "{'loss': 1.0441, 'learning_rate': 1.0466666666666668e-05, 'epoch': 0.06}        \r\n",
+      "{'loss': 1.2198, 'learning_rate': 1.04e-05, 'epoch': 0.06}                      \r\n",
+      "{'loss': 0.019, 'learning_rate': 1.0333333333333335e-05, 'epoch': 0.06}         \r\n",
+      "{'loss': 0.0127, 'learning_rate': 1.0266666666666668e-05, 'epoch': 0.06}        \r\n",
+      "{'loss': 0.0606, 'learning_rate': 1.02e-05, 'epoch': 0.06}                      \r\n",
+      "{'loss': 0.1299, 'learning_rate': 1.0133333333333335e-05, 'epoch': 0.06}        \r\n",
+      "{'loss': 0.0678, 'learning_rate': 1.0066666666666666e-05, 'epoch': 0.06}        \r\n",
+      "{'loss': 0.0037, 'learning_rate': 1e-05, 'epoch': 0.06}                         \r\n",
+      "{'loss': 0.4082, 'learning_rate': 9.933333333333334e-06, 'epoch': 0.06}         \r\n",
+      "{'loss': 0.1784, 'learning_rate': 9.866666666666668e-06, 'epoch': 0.06}         \r\n",
+      "{'loss': 0.0074, 'learning_rate': 9.800000000000001e-06, 'epoch': 0.06}         \r\n",
+      "{'loss': 0.148, 'learning_rate': 9.733333333333334e-06, 'epoch': 0.06}          \r\n",
+      "{'loss': 0.0005, 'learning_rate': 9.666666666666667e-06, 'epoch': 0.06}         \r\n",
+      "{'loss': 0.3039, 'learning_rate': 9.600000000000001e-06, 'epoch': 0.06}         \r\n",
+      "{'loss': 0.0024, 'learning_rate': 9.533333333333334e-06, 'epoch': 0.06}         \r\n",
+      "{'loss': 0.2657, 'learning_rate': 9.466666666666667e-06, 'epoch': 0.06}         \r\n",
+      "{'loss': 0.0117, 'learning_rate': 9.4e-06, 'epoch': 0.06}                       \r\n",
+      "{'loss': 0.001, 'learning_rate': 9.333333333333334e-06, 'epoch': 0.06}          \r\n",
+      "{'loss': 0.0045, 'learning_rate': 9.266666666666667e-06, 'epoch': 0.06}         \r\n",
+      "{'loss': 0.0013, 'learning_rate': 9.200000000000002e-06, 'epoch': 0.06}         \r\n",
+      "{'loss': 0.4838, 'learning_rate': 9.133333333333335e-06, 'epoch': 0.07}         \r\n",
+      "{'loss': 0.164, 'learning_rate': 9.066666666666667e-06, 'epoch': 0.07}          \r\n",
+      "{'loss': 0.0084, 'learning_rate': 9e-06, 'epoch': 0.07}                         \r\n",
+      "{'loss': 0.0207, 'learning_rate': 8.933333333333333e-06, 'epoch': 0.07}         \r\n",
+      "{'loss': 0.3102, 'learning_rate': 8.866666666666668e-06, 'epoch': 0.07}         \r\n",
+      "{'loss': 0.0723, 'learning_rate': 8.8e-06, 'epoch': 0.07}                       \r\n",
+      "{'loss': 0.0009, 'learning_rate': 8.733333333333333e-06, 'epoch': 0.07}         \r\n",
+      "{'loss': 0.1778, 'learning_rate': 8.666666666666668e-06, 'epoch': 0.07}         \r\n",
+      "{'loss': 1.0444, 'learning_rate': 8.6e-06, 'epoch': 0.07}                       \r\n",
+      "{'loss': 0.1915, 'learning_rate': 8.533333333333335e-06, 'epoch': 0.07}         \r\n",
+      "{'loss': 0.0191, 'learning_rate': 8.466666666666668e-06, 'epoch': 0.07}         \r\n",
+      "{'loss': 0.0044, 'learning_rate': 8.400000000000001e-06, 'epoch': 0.07}         \r\n",
+      "{'loss': 0.5965, 'learning_rate': 8.333333333333334e-06, 'epoch': 0.07}         \r\n",
+      "{'loss': 0.0201, 'learning_rate': 8.266666666666667e-06, 'epoch': 0.07}         \r\n",
+      "{'loss': 1.1224, 'learning_rate': 8.2e-06, 'epoch': 0.07}                       \r\n",
+      "{'loss': 0.0041, 'learning_rate': 8.133333333333334e-06, 'epoch': 0.07}         \r\n",
+      "{'loss': 0.0071, 'learning_rate': 8.066666666666667e-06, 'epoch': 0.07}         \r\n",
+      "{'loss': 0.0885, 'learning_rate': 8.000000000000001e-06, 'epoch': 0.07}         \r\n",
+      "{'loss': 0.394, 'learning_rate': 7.933333333333334e-06, 'epoch': 0.07}          \r\n",
+      "{'loss': 0.0002, 'learning_rate': 7.866666666666667e-06, 'epoch': 0.07}         \r\n",
+      "{'loss': 0.2249, 'learning_rate': 7.800000000000002e-06, 'epoch': 0.07}         \r\n",
+      "{'loss': 0.2116, 'learning_rate': 7.733333333333334e-06, 'epoch': 0.07}         \r\n",
+      "{'loss': 0.0159, 'learning_rate': 7.666666666666667e-06, 'epoch': 0.07}         \r\n",
+      "{'loss': 0.0027, 'learning_rate': 7.600000000000001e-06, 'epoch': 0.07}         \r\n",
+      "{'loss': 0.1615, 'learning_rate': 7.533333333333334e-06, 'epoch': 0.07}         \r\n",
+      "{'loss': 0.0401, 'learning_rate': 7.4666666666666675e-06, 'epoch': 0.08}        \r\n",
+      "{'loss': 0.0089, 'learning_rate': 7.4e-06, 'epoch': 0.08}                       \r\n",
+      "{'loss': 0.0712, 'learning_rate': 7.333333333333333e-06, 'epoch': 0.08}         \r\n",
+      "{'loss': 0.5022, 'learning_rate': 7.266666666666668e-06, 'epoch': 0.08}         \r\n",
+      "{'loss': 0.1132, 'learning_rate': 7.2000000000000005e-06, 'epoch': 0.08}        \r\n",
+      "{'loss': 0.3318, 'learning_rate': 7.133333333333334e-06, 'epoch': 0.08}         \r\n",
+      "{'loss': 0.0071, 'learning_rate': 7.066666666666667e-06, 'epoch': 0.08}         \r\n",
+      "{'loss': 0.6094, 'learning_rate': 7e-06, 'epoch': 0.08}                         \r\n",
+      "{'loss': 0.0022, 'learning_rate': 6.9333333333333344e-06, 'epoch': 0.08}        \r\n",
+      "{'loss': 0.4046, 'learning_rate': 6.866666666666667e-06, 'epoch': 0.08}         \r\n",
+      "{'loss': 0.3455, 'learning_rate': 6.800000000000001e-06, 'epoch': 0.08}         \r\n",
+      "{'loss': 0.5033, 'learning_rate': 6.733333333333334e-06, 'epoch': 0.08}         \r\n",
+      "{'loss': 0.1898, 'learning_rate': 6.666666666666667e-06, 'epoch': 0.08}         \r\n",
+      " 67%|███████████████████████████▎             | 200/300 [05:00<02:30,  1.51s/it][INFO|tokenization_utils_base.py:2428] 2023-11-24 13:49:25,887 >> tokenizer config file saved in output/news-20231124-134412-2e-05/checkpoint-200/tokenizer_config.json\r\n",
+      "[INFO|tokenization_utils_base.py:2437] 2023-11-24 13:49:25,887 >> Special tokens file saved in output/news-20231124-134412-2e-05/checkpoint-200/special_tokens_map.json\r\n",
+      "{'loss': 0.2219, 'learning_rate': 6.600000000000001e-06, 'epoch': 0.08}         \r\n",
+      "{'loss': 0.0261, 'learning_rate': 6.533333333333334e-06, 'epoch': 0.08}         \r\n",
+      "{'loss': 0.0268, 'learning_rate': 6.466666666666667e-06, 'epoch': 0.08}         \r\n",
+      "{'loss': 0.2255, 'learning_rate': 6.4000000000000006e-06, 'epoch': 0.08}        \r\n",
+      "{'loss': 0.0023, 'learning_rate': 6.333333333333333e-06, 'epoch': 0.08}         \r\n",
+      "{'loss': 0.003, 'learning_rate': 6.266666666666668e-06, 'epoch': 0.08}          \r\n",
+      "{'loss': 0.0272, 'learning_rate': 6.200000000000001e-06, 'epoch': 0.08}         \r\n",
+      "{'loss': 0.0585, 'learning_rate': 6.133333333333334e-06, 'epoch': 0.08}         \r\n",
+      "{'loss': 0.0094, 'learning_rate': 6.066666666666667e-06, 'epoch': 0.08}         \r\n",
+      "{'loss': 0.163, 'learning_rate': 6e-06, 'epoch': 0.08}                          \r\n",
+      "{'loss': 0.0036, 'learning_rate': 5.933333333333335e-06, 'epoch': 0.08}         \r\n",
+      "{'loss': 0.0781, 'learning_rate': 5.8666666666666675e-06, 'epoch': 0.08}        \r\n",
+      "{'loss': 0.0089, 'learning_rate': 5.8e-06, 'epoch': 0.09}                       \r\n",
+      "{'loss': 0.5827, 'learning_rate': 5.733333333333334e-06, 'epoch': 0.09}         \r\n",
+      "{'loss': 0.0053, 'learning_rate': 5.666666666666667e-06, 'epoch': 0.09}         \r\n",
+      "{'loss': 0.0179, 'learning_rate': 5.600000000000001e-06, 'epoch': 0.09}         \r\n",
+      "{'loss': 1.8615, 'learning_rate': 5.533333333333334e-06, 'epoch': 0.09}         \r\n",
+      "{'loss': 0.3423, 'learning_rate': 5.466666666666667e-06, 'epoch': 0.09}         \r\n",
+      "{'loss': 0.5502, 'learning_rate': 5.400000000000001e-06, 'epoch': 0.09}         \r\n",
+      "{'loss': 0.1772, 'learning_rate': 5.333333333333334e-06, 'epoch': 0.09}         \r\n",
+      "{'loss': 0.075, 'learning_rate': 5.2666666666666665e-06, 'epoch': 0.09}         \r\n",
+      "{'loss': 0.0059, 'learning_rate': 5.2e-06, 'epoch': 0.09}                       \r\n",
+      "{'loss': 0.0042, 'learning_rate': 5.133333333333334e-06, 'epoch': 0.09}         \r\n",
+      "{'loss': 0.0213, 'learning_rate': 5.0666666666666676e-06, 'epoch': 0.09}        \r\n",
+      "{'loss': 0.0174, 'learning_rate': 5e-06, 'epoch': 0.09}                         \r\n",
+      "{'loss': 1.0351, 'learning_rate': 4.933333333333334e-06, 'epoch': 0.09}         \r\n",
+      "{'loss': 0.0026, 'learning_rate': 4.866666666666667e-06, 'epoch': 0.09}         \r\n",
+      "{'loss': 0.0026, 'learning_rate': 4.800000000000001e-06, 'epoch': 0.09}         \r\n",
+      "{'loss': 0.5251, 'learning_rate': 4.7333333333333335e-06, 'epoch': 0.09}        \r\n",
+      "{'loss': 0.2601, 'learning_rate': 4.666666666666667e-06, 'epoch': 0.09}         \r\n",
+      "{'loss': 0.006, 'learning_rate': 4.600000000000001e-06, 'epoch': 0.09}          \r\n",
+      "{'loss': 0.8758, 'learning_rate': 4.533333333333334e-06, 'epoch': 0.09}         \r\n",
+      "{'loss': 0.2939, 'learning_rate': 4.4666666666666665e-06, 'epoch': 0.09}        \r\n",
+      "{'loss': 0.0141, 'learning_rate': 4.4e-06, 'epoch': 0.09}                       \r\n",
+      "{'loss': 0.001, 'learning_rate': 4.333333333333334e-06, 'epoch': 0.09}          \r\n",
+      "{'loss': 0.0176, 'learning_rate': 4.266666666666668e-06, 'epoch': 0.09}         \r\n",
+      "{'loss': 0.4934, 'learning_rate': 4.2000000000000004e-06, 'epoch': 0.09}        \r\n",
+      "{'loss': 0.3413, 'learning_rate': 4.133333333333333e-06, 'epoch': 0.1}          \r\n",
+      "{'loss': 0.0059, 'learning_rate': 4.066666666666667e-06, 'epoch': 0.1}          \r\n",
+      "{'loss': 0.0379, 'learning_rate': 4.000000000000001e-06, 'epoch': 0.1}          \r\n",
+      "{'loss': 0.7871, 'learning_rate': 3.9333333333333335e-06, 'epoch': 0.1}         \r\n",
+      "{'loss': 0.0035, 'learning_rate': 3.866666666666667e-06, 'epoch': 0.1}          \r\n",
+      "{'loss': 0.3481, 'learning_rate': 3.8000000000000005e-06, 'epoch': 0.1}         \r\n",
+      "{'loss': 0.1827, 'learning_rate': 3.7333333333333337e-06, 'epoch': 0.1}         \r\n",
+      "{'loss': 0.4669, 'learning_rate': 3.6666666666666666e-06, 'epoch': 0.1}         \r\n",
+      "{'loss': 0.8078, 'learning_rate': 3.6000000000000003e-06, 'epoch': 0.1}         \r\n",
+      "{'loss': 1.1314, 'learning_rate': 3.5333333333333335e-06, 'epoch': 0.1}         \r\n",
+      "{'loss': 0.0124, 'learning_rate': 3.4666666666666672e-06, 'epoch': 0.1}         \r\n",
+      "{'loss': 0.0116, 'learning_rate': 3.4000000000000005e-06, 'epoch': 0.1}         \r\n",
+      "{'loss': 0.2296, 'learning_rate': 3.3333333333333333e-06, 'epoch': 0.1}         \r\n",
+      "{'loss': 0.0037, 'learning_rate': 3.266666666666667e-06, 'epoch': 0.1}          \r\n",
+      "{'loss': 0.0882, 'learning_rate': 3.2000000000000003e-06, 'epoch': 0.1}         \r\n",
+      "{'loss': 0.0073, 'learning_rate': 3.133333333333334e-06, 'epoch': 0.1}          \r\n",
+      "{'loss': 0.013, 'learning_rate': 3.066666666666667e-06, 'epoch': 0.1}           \r\n",
+      "{'loss': 0.0027, 'learning_rate': 3e-06, 'epoch': 0.1}                          \r\n",
+      "{'loss': 0.7103, 'learning_rate': 2.9333333333333338e-06, 'epoch': 0.1}         \r\n",
+      "{'loss': 0.0022, 'learning_rate': 2.866666666666667e-06, 'epoch': 0.1}          \r\n",
+      "{'loss': 0.4343, 'learning_rate': 2.8000000000000003e-06, 'epoch': 0.1}         \r\n",
+      "{'loss': 1.1652, 'learning_rate': 2.7333333333333336e-06, 'epoch': 0.1}         \r\n",
+      "{'loss': 0.377, 'learning_rate': 2.666666666666667e-06, 'epoch': 0.1}           \r\n",
+      "{'loss': 0.0025, 'learning_rate': 2.6e-06, 'epoch': 0.1}                        \r\n",
+      "{'loss': 0.072, 'learning_rate': 2.5333333333333338e-06, 'epoch': 0.1}          \r\n",
+      "{'loss': 0.0035, 'learning_rate': 2.466666666666667e-06, 'epoch': 0.11}         \r\n",
+      "{'loss': 0.0176, 'learning_rate': 2.4000000000000003e-06, 'epoch': 0.11}        \r\n",
+      "{'loss': 0.0404, 'learning_rate': 2.3333333333333336e-06, 'epoch': 0.11}        \r\n",
+      "{'loss': 0.0243, 'learning_rate': 2.266666666666667e-06, 'epoch': 0.11}         \r\n",
+      "{'loss': 0.0197, 'learning_rate': 2.2e-06, 'epoch': 0.11}                       \r\n",
+      "{'loss': 0.0558, 'learning_rate': 2.133333333333334e-06, 'epoch': 0.11}         \r\n",
+      "{'loss': 0.0396, 'learning_rate': 2.0666666666666666e-06, 'epoch': 0.11}        \r\n",
+      "{'loss': 0.0042, 'learning_rate': 2.0000000000000003e-06, 'epoch': 0.11}        \r\n",
+      "{'loss': 0.2281, 'learning_rate': 1.9333333333333336e-06, 'epoch': 0.11}        \r\n",
+      "{'loss': 0.0594, 'learning_rate': 1.8666666666666669e-06, 'epoch': 0.11}        \r\n",
+      "{'loss': 0.0016, 'learning_rate': 1.8000000000000001e-06, 'epoch': 0.11}        \r\n",
+      "{'loss': 0.1489, 'learning_rate': 1.7333333333333336e-06, 'epoch': 0.11}        \r\n",
+      "{'loss': 0.0053, 'learning_rate': 1.6666666666666667e-06, 'epoch': 0.11}        \r\n",
+      "{'loss': 0.0131, 'learning_rate': 1.6000000000000001e-06, 'epoch': 0.11}        \r\n",
+      "{'loss': 0.0517, 'learning_rate': 1.5333333333333334e-06, 'epoch': 0.11}        \r\n",
+      "{'loss': 2.0054, 'learning_rate': 1.4666666666666669e-06, 'epoch': 0.11}        \r\n",
+      "{'loss': 0.863, 'learning_rate': 1.4000000000000001e-06, 'epoch': 0.11}         \r\n",
+      "{'loss': 0.0038, 'learning_rate': 1.3333333333333334e-06, 'epoch': 0.11}        \r\n",
+      "{'loss': 0.0003, 'learning_rate': 1.2666666666666669e-06, 'epoch': 0.11}        \r\n",
+      "{'loss': 0.0553, 'learning_rate': 1.2000000000000002e-06, 'epoch': 0.11}        \r\n",
+      "{'loss': 0.0076, 'learning_rate': 1.1333333333333334e-06, 'epoch': 0.11}        \r\n",
+      "{'loss': 0.0481, 'learning_rate': 1.066666666666667e-06, 'epoch': 0.11}         \r\n",
+      "{'loss': 0.5921, 'learning_rate': 1.0000000000000002e-06, 'epoch': 0.11}        \r\n",
+      "{'loss': 0.3559, 'learning_rate': 9.333333333333334e-07, 'epoch': 0.11}         \r\n",
+      "{'loss': 0.0014, 'learning_rate': 8.666666666666668e-07, 'epoch': 0.11}         \r\n",
+      "{'loss': 0.4355, 'learning_rate': 8.000000000000001e-07, 'epoch': 0.12}         \r\n",
+      "{'loss': 0.0764, 'learning_rate': 7.333333333333334e-07, 'epoch': 0.12}         \r\n",
+      "{'loss': 0.0035, 'learning_rate': 6.666666666666667e-07, 'epoch': 0.12}         \r\n",
+      "{'loss': 0.0042, 'learning_rate': 6.000000000000001e-07, 'epoch': 0.12}         \r\n",
+      "{'loss': 0.0086, 'learning_rate': 5.333333333333335e-07, 'epoch': 0.12}         \r\n",
+      "{'loss': 0.2114, 'learning_rate': 4.666666666666667e-07, 'epoch': 0.12}         \r\n",
+      "{'loss': 0.223, 'learning_rate': 4.0000000000000003e-07, 'epoch': 0.12}         \r\n",
+      "{'loss': 0.0044, 'learning_rate': 3.3333333333333335e-07, 'epoch': 0.12}        \r\n",
+      "{'loss': 0.6593, 'learning_rate': 2.666666666666667e-07, 'epoch': 0.12}         \r\n",
+      "{'loss': 0.1374, 'learning_rate': 2.0000000000000002e-07, 'epoch': 0.12}        \r\n",
+      "{'loss': 0.1037, 'learning_rate': 1.3333333333333336e-07, 'epoch': 0.12}        \r\n",
+      "{'loss': 0.0181, 'learning_rate': 6.666666666666668e-08, 'epoch': 0.12}         \r\n",
+      "{'loss': 0.0783, 'learning_rate': 0.0, 'epoch': 0.12}                           \r\n",
+      "100%|█████████████████████████████████████████| 300/300 [07:31<00:00,  1.51s/it][INFO|tokenization_utils_base.py:2428] 2023-11-24 13:51:56,849 >> tokenizer config file saved in output/news-20231124-134412-2e-05/checkpoint-300/tokenizer_config.json\r\n",
+      "[INFO|tokenization_utils_base.py:2437] 2023-11-24 13:51:56,849 >> Special tokens file saved in output/news-20231124-134412-2e-05/checkpoint-300/special_tokens_map.json\r\n",
+      "[INFO|trainer.py:1955] 2023-11-24 13:51:56,875 >> \r\n",
+      "\r\n",
+      "Training completed. Do not forget to share your model on huggingface.co/models =)\r\n",
+      "\r\n",
+      "\r\n",
+      "{'train_runtime': 451.0866, 'train_samples_per_second': 1.33, 'train_steps_per_second': 0.665, 'train_loss': 0.2702767427762349, 'epoch': 0.12}\r\n",
+      "100%|█████████████████████████████████████████| 300/300 [07:31<00:00,  1.50s/it]\r\n",
+      "[INFO|tokenization_utils_base.py:2428] 2023-11-24 13:51:56,887 >> tokenizer config file saved in output/news-20231124-134412-2e-05/tokenizer_config.json\r\n",
+      "[INFO|tokenization_utils_base.py:2437] 2023-11-24 13:51:56,887 >> Special tokens file saved in output/news-20231124-134412-2e-05/special_tokens_map.json\r\n"
+     ]
+    }
+   ],
+   "source": [
+    "!which python\n",
+    "import os\n",
+    "from datetime import datetime\n",
+    "import random\n",
+    "\n",
+    "# 定义变量\n",
+    "lr = 2e-5\n",
+    "num_gpus = 1\n",
+    "lora_rank = 8\n",
+    "lora_alpha = 32\n",
+    "lora_dropout = 0.1\n",
+    "max_source_len = 512\n",
+    "max_target_len = 128\n",
+    "dev_batch_size = 1\n",
+    "grad_accumularion_steps = 2\n",
+    "max_step = 300\n",
+    "save_interval = 100\n",
+    "max_seq_len = 512\n",
+    "logging_steps=1\n",
+    "\n",
+    "run_name = \"news\"\n",
+    "dataset_path = \"data/toutiao_cat_data_example.jsonl\"\n",
+    "datestr = datetime.now().strftime(\"%Y%m%d-%H%M%S\")\n",
+    "output_dir = f\"output/{run_name}-{datestr}-{lr}\"\n",
+    "master_port = random.randint(10000, 65535)\n",
+    "\n",
+    "os.makedirs(output_dir, exist_ok=True)\n",
+    "# 构建命令\n",
+    "command = f\"\"\"\n",
+    "torchrun --standalone --nnodes=1 --nproc_per_node={num_gpus} ../finetune_basemodel_demo/finetune.py \\\n",
+    "    --train_format input-output \\\n",
+    "    --train_file {dataset_path} \\\n",
+    "    --lora_rank {lora_rank} \\\n",
+    "    --lora_alpha {lora_alpha} \\\n",
+    "    --lora_dropout {lora_dropout} \\\n",
+    "    --max_seq_length {max_seq_len} \\\n",
+    "    --preprocessing_num_workers 1 \\\n",
+    "    --model_name_or_path {model_path} \\\n",
+    "    --output_dir {output_dir} \\\n",
+    "    --per_device_train_batch_size 1 \\\n",
+    "    --gradient_accumulation_steps 2 \\\n",
+    "    --max_steps {max_step} \\\n",
+    "    --logging_steps {logging_steps} \\\n",
+    "    --save_steps {save_interval} \\\n",
+    "    --learning_rate {lr}\n",
+    "\"\"\"\n",
+    "\n",
+    "# 在 Notebook 中执行命令\n",
+    "!{command}"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-11-24T13:44:12.468751Z",
+     "end_time": "2023-11-24T13:51:59.534815Z"
+    }
+   }
+  },
+  {
+   "cell_type": "markdown",
+   "source": [
+    "## 使用微调的模型进行推理预测\n",
+    "现在，我们已经完成了模型的微调，接下来，我们将使用微调后的模型进行推理。我们使用与微调时相同的提示词，并使用一些没有出现的模型效果来复现推理结果。"
+   ],
+   "metadata": {
+    "collapsed": false
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "Loading checkpoint shards:   0%|          | 0/7 [00:00<?, ?it/s]",
+      "application/vnd.jupyter.widget-view+json": {
+       "version_major": 2,
+       "version_minor": 0,
+       "model_id": "21b29455e65c4540886188d1fd5d68aa"
+      }
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "import torch\n",
+    "from transformers import AutoModel, AutoTokenizer\n",
+    "import os\n",
+    "from peft import get_peft_model, LoraConfig, TaskType\n",
+    "\n",
+    "# 参数设置\n",
+    "lora_path = output_dir + \"pytorch_model.bin\"\n",
+    "\n",
+    "# 加载分词器和模型\n",
+    "tokenizer = AutoTokenizer.from_pretrained(tokenizer_path, trust_remote_code=True)\n",
+    "model = AutoModel.from_pretrained(model_path, load_in_8bit=False, trust_remote_code=True).to(device)\n",
+    "\n",
+    "# LoRA 模型配置\n",
+    "peft_config = LoraConfig(\n",
+    "    task_type=TaskType.CAUSAL_LM, inference_mode=True,\n",
+    "    target_modules=['query_key_value'],\n",
+    "    r=8, lora_alpha=32, lora_dropout=0.1\n",
+    ")\n",
+    "model = get_peft_model(model, peft_config)\n",
+    "if os.path.exists(lora_path):\n",
+    "    model.load_state_dict(torch.load(lora_path), strict=False)\n"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-11-24T13:51:59.522046Z",
+     "end_time": "2023-11-24T13:52:12.130299Z"
+    }
+   }
+  },
+  {
+   "cell_type": "markdown",
+   "source": [
+    "我们使用同样的提示词进行推理。"
+   ],
+   "metadata": {
+    "collapsed": false
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "outputs": [],
+   "source": [
+    "inputs = tokenizer(PROMPT, return_tensors=\"pt\").to(device)\n",
+    "response = model.generate(input_ids=inputs[\"input_ids\"],max_new_tokens=max_new_tokens,temperature=temperature,top_p=top_p,do_sample=True)\n",
+    "response = response[0, inputs[\"input_ids\"].shape[-1]:]\n",
+    "response = tokenizer.decode(response, skip_special_tokens=True)"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-11-24T13:52:12.132308Z",
+     "end_time": "2023-11-24T13:52:12.296837Z"
+    }
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "outputs": [
+    {
+     "data": {
+      "text/plain": "'news_tech'"
+     },
+     "execution_count": 9,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "response"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-11-24T13:52:12.296837Z",
+     "end_time": "2023-11-24T13:52:12.297910Z"
+    }
+   }
+  },
+  {
+   "cell_type": "markdown",
+   "source": [
+    "这一次，模型成功给出了理想的答案。我们结束实操训练，删除模型并释放显存。"
+   ],
+   "metadata": {
+    "collapsed": false
+   }
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "outputs": [],
+   "source": [
+    "del model\n",
+    "torch.cuda.empty_cache()"
+   ],
+   "metadata": {
+    "collapsed": false,
+    "ExecuteTime": {
+     "start_time": "2023-11-24T13:54:07.355796Z",
+     "end_time": "2023-11-24T13:54:07.507254Z"
+    }
+   }
+  },
+  {
+   "cell_type": "markdown",
+   "source": [
+    "## 总结\n",
+    "在本实践手册中，我们让开发者体验了 `ChatGLM3-6B` 模型在经过微调前后在 `新闻标题分类` 任务中表现。\n",
+    "我们可以发现：\n",
+    "\n",
+    "在具有混淆的新闻分类中，原始的模型可能受到了误导，不能有效的进行分类，而经过简单微调后的模型，已经具备了正确分类的能力。\n",
+    "因此，对于有更高要求的专业分类任务，我们可以使用微调的方式对模型进行简单微调，实现更好的任务完成效果。\n"
+   ],
+   "metadata": {
+    "collapsed": false
+   }
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 2
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython2",
+   "version": "2.7.6"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 0
+}
diff --git a/finetune_basemodel_demo/README.md b/finetune_basemodel_demo/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..8c135b723101229389418ef325587bf3968437d3
--- /dev/null
+++ b/finetune_basemodel_demo/README.md
@@ -0,0 +1,90 @@
+# ChatGLM3-6B-base 微调示例
+
+本目录提供 ChatGLM3-6B-base 模型的微调示例，目前，仅包含了Lora微调。
+
+如果将模型下载到了本地，本文和代码中的 `THUDM/chatglm3-6b-base` 字段均应替换为相应地址以从本地加载模型。
+
+运行示例需要 `python>=3.10`，除基础的 `torch` 依赖外，示例代码运行还需要依赖 
+
+```bash
+pip install -r requirements.txt
+```
+## 多轮对话格式
+
+`base`模型不具备对话能力，仅能够生成单轮回复。如果你希望使用多轮对话模型，使用`Chat`模型进行微调。
+
+
+## 数据集要求
+
+格式上，请使用`alpaca`数据集。
+
+```bash
+
+{"context": "hello", "target": "hi,I am ChatGLM3"}
+
+```
+其中，`context`是对话的上文，也就是模型的输入，`target`是对话的下文，也就是模型的输出。
+
+## 微调模型
+
+以下脚本提供了微调模型的参考方式。
+```bash
+./scripts/finetune_lora.sh  # 使用Lora微调
+```
+如果你使用的是windows系统，请将脚本改为`bat`文件执行。同时，如果你是单卡运行环境，也可以将脚本中中的这行代码
+```
+torchrun --standalone --nnodes=1 --nproc_per_node=$NUM_GPUS finetune.py 
+```
+修改为
+```
+python finetune.py 
+```
+请注意，这样修改后，你需要手动设置`CUDA_VISIBLE_DEVICES`环境变量，以控制使用哪张显卡。
+
+### 提示
+
+1. 微调代码在开始训练前，会先打印首条训练数据的预处理信息，显示为
+
+    ```log
+    Sanity Check >>>>>>>>>>>>>
+             '[gMASK]':  64790 ->   -100
+                 'sop':  64792 ->   -100
+          '<|system|>':  64794 ->   -100
+                    '':  30910 ->   -100
+                  '\n':     13 ->   -100
+              'Answer':  20115 ->   -100
+                 'the':    267 ->   -100
+           'following':   1762 ->   -100
+                      ...
+                'know':    683 ->   -100
+                 'the':    267 ->   -100
+            'response':   3010 ->   -100
+             'details':   3296 ->   -100
+                   '.':  30930 ->   -100
+       '<|assistant|>':  64796 ->   -100
+                    '':  30910 ->  30910
+                  '\n':     13 ->     13
+                   'I':    307 ->    307
+                'need':    720 ->    720
+                  'to':    289 ->    289
+                 'use':    792 ->    792
+                      ...
+                    '':      0 ->   -100
+                    '':      0 ->   -100 （有若干个）
+    <<<<<<<<<<<<< Sanity Check
+    ```
+    字样，每行依次表示一个 detokenized string, token_id 和 target_id。可在日志中查看这部分的 `loss_mask` 是否符合预期。若不符合，可能需要调整代码或数据。
+2. 参考显存用量
+   - 按照官方脚本的默认参数运行，每一张显卡占用显存为 `23GB`。
+3. 若尝试后发现显存不足，可以考虑
+    - 尝试降低 `DEV_BATCH_SIZE` 并提升 `GRAD_ACCUMULARION_STEPS`
+    - 尝试降低 `MAX_SEQ_LEN`，但是这可能会影响模型的性能
+4. 数据量和 loss 衰减
+    - 训练中 loss 在数个 step 中回升属于正常现象，但是如果连续不降，则需要检查数据集是否存在问题（覆盖范围过广，相关性太低等）。你也可以选择降低 `LEARNING_RATE` 或者提升 `WARMUP_STEPS`。
+    - 通常来说，loss值在 `0.5` 以下时，就属于比较好的状态。
+5. 训练数据和轮次
+    - `Max Steps` 为训练执行的步骤，与数据集的数量无关，这个参数不等于`Epoch`。
+    - `Batch Size` 为每个GPU的batch size，如果你使用了多卡训练，那么实际的batch size为 `Batch Size * GPU数量`。
+
+## 注意事项
++ 基座模型不支持对话，工具，代码生成等能力，仅支持文本生成。如果你需要对话能力，请使用`Chat`模型和对应的微调框架。
\ No newline at end of file
diff --git a/finetune_basemodel_demo/arguments.py b/finetune_basemodel_demo/arguments.py
new file mode 100644
index 0000000000000000000000000000000000000000..28d0c87eac96683c22760f6ea35ad3422f195d66
--- /dev/null
+++ b/finetune_basemodel_demo/arguments.py
@@ -0,0 +1,182 @@
+from dataclasses import dataclass, field
+from typing import Optional
+
+
+@dataclass
+class ModelArguments:
+    """
+    Arguments pertaining to which model/config/tokenizer we are going to fine-tune from.
+    """
+
+    model_name_or_path: str = field(
+        metadata={"help": "Path to pretrained model or model identifier from huggingface.co/models"}
+    )
+
+    lora_checkpoint: str = field(
+        default=None, metadata={"help": "Path to lora checkpoints"}
+    )
+    config_name: Optional[str] = field(
+        default=None, metadata={"help": "Pretrained config name or path if not the same as model_name"}
+    )
+    tokenizer_name: Optional[str] = field(
+        default=None, metadata={"help": "Pretrained tokenizer name or path if not the same as model_name"}
+    )
+    cache_dir: Optional[str] = field(
+        default=None,
+        metadata={"help": "Where to store the pretrained models downloaded from huggingface.co"},
+    )
+    use_fast_tokenizer: bool = field(
+        default=True,
+        metadata={"help": "Whether to use one of the fast tokenizer (backed by the tokenizers library) or not."},
+    )
+    model_revision: str = field(
+        default="main",
+        metadata={"help": "The specific model version to use (can be a branch name, tag name or commit id)."},
+    )
+    use_auth_token: bool = field(
+        default=False,
+        metadata={
+            "help": (
+                "Will use the token generated when running `huggingface-cli login` (necessary to use this script "
+                "with private models)."
+            )
+        },
+    )
+    resize_position_embeddings: Optional[bool] = field(
+        default=None,
+        metadata={
+            "help": (
+                "Whether to automatically resize the position embeddings if `max_source_length` exceeds "
+                "the model's position embeddings."
+            )
+        },
+    )
+    quantization_bit: Optional[int] = field(
+        default=None,
+        metadata={
+            "help": (
+                "An optional parameter specifying the number of bits used for quantization. "
+                "Quantization is a process that reduces the model size by limiting the number of "
+                "bits that represent each weight in the model. A lower number of bits can reduce "
+                "the model size and speed up inference, but might also decrease model accuracy. "
+                "If not set (None), quantization is not applied."
+            )
+        },
+    )
+
+    lora_rank: Optional[int] = field(
+        default=8,
+        metadata={
+            "help": (
+                "balancing between complexity and model flexibility. A higher rank allows more "
+                "complex adaptations but increases the number of parameters and computational cost."
+            )
+        },
+    )
+    lora_alpha: Optional[float] = field(
+        default=32,
+        metadata={
+            "help": (
+                "A higher value results in more significant adjustments, potentially improving adaptation to new tasks or data, "
+                "but might also risk overfitting. A lower value makes smaller adjustments, possibly maintaining better generalization."
+            )
+        }, )
+
+    lora_dropout: Optional[float] = field(
+        default=0.1,
+        metadata={
+            "help": (
+                "during training to prevent the model from overly relying on specific patterns in the training data. "
+                "Higher dropout rates can improve model generalization but may reduce learning efficiency."
+            )
+        },
+    )
+
+
+@dataclass
+class DataTrainingArguments:
+    """
+    Arguments pertaining to what data we are going to input our model for training and eval.
+    """
+    train_file: Optional[str] = field(
+        default=None, metadata={"help": "The input training data file (a jsonlines or csv file)."}
+    )
+
+    max_seq_length: Optional[int] = field(
+        default=2048,
+        metadata={
+            "help": (
+                "The maximum total input sequence length after tokenization. Sequences longer "
+                "than this will be truncated."
+            )
+        },
+    )
+
+    max_source_length: Optional[int] = field(
+        default=1024,
+        metadata={
+            "help": (
+                "The maximum total input sequence length after tokenization. Sequences longer "
+                "than this will be truncated, sequences shorter will be padded."
+            )
+        },
+    )
+    max_target_length: Optional[int] = field(
+        default=128,
+        metadata={
+            "help": (
+                "The maximum total sequence length for target text after tokenization. Sequences longer "
+                "than this will be truncated, sequences shorter will be padded."
+            )
+        },
+    )
+
+    train_format: str = field(
+        default=None, metadata={"help": "The format of the training data file (mulit-turn or input-output)"},
+    )
+
+    overwrite_cache: bool = field(
+        default=False, metadata={"help": "Overwrite the cached training and evaluation sets"}
+    )
+
+    preprocessing_num_workers: Optional[int] = field(
+        default=None,
+        metadata={"help": "The number of processes to use for the preprocessing."},
+    )
+
+    max_seq_length: Optional[int] = field(
+        default=1024,
+        metadata={
+            "help": (
+                "The maximum total input sequence length after tokenization. Sequences longer "
+                "than this will be truncated, sequences shorter will be padded."
+            )
+        },
+    )
+
+    pad_to_max_length: bool = field(
+        default=False,
+        metadata={
+            "help": (
+                "Whether to pad all samples to model maximum sentence length. "
+                "If False, will pad the samples dynamically when batching to the maximum length in the batch. More "
+                "efficient on GPU but very bad for TPU."
+            )
+        },
+    )
+
+    max_train_samples: Optional[int] = field(
+        default=None,
+        metadata={
+            "help": (
+                "For debugging purposes or quicker training, truncate the number of training examples to this "
+                "value if set."
+            )
+        },
+    )
+
+    def __post_init__(self):
+        extension = self.train_file.split(".")[-1]
+        assert extension in {"jsonl", "json"}, "`train_file` should be a jsonl or a json file."
+
+        assert self.train_format in {"multi-turn", "input-output"}
diff --git a/finetune_basemodel_demo/finetune.py b/finetune_basemodel_demo/finetune.py
new file mode 100644
index 0000000000000000000000000000000000000000..cd827948b2dbe47670b086dfef1d6d3566314739
--- /dev/null
+++ b/finetune_basemodel_demo/finetune.py
@@ -0,0 +1,165 @@
+#!/usr/bin/env python
+# coding=utf-8
+# Copyright 2021 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""
+Fine-tuning the library models for sequence to sequence.
+"""
+# You can also adapt this script on your own sequence to sequence task. Pointers for this are left as comments.
+# Adapted from
+
+
+import logging
+import os
+import sys
+import torch
+import json
+import transformers
+from transformers import (
+    AutoModel,
+    AutoTokenizer,
+    DataCollatorForSeq2Seq,
+    HfArgumentParser,
+    Seq2SeqTrainingArguments,
+    set_seed,
+)
+from trainer import LoRATrainer
+from arguments import ModelArguments, DataTrainingArguments
+from peft import get_peft_model, LoraConfig, TaskType, prepare_model_for_kbit_training
+from preprocess_utils import sanity_check, InputOutputDataset
+
+
+logger = logging.getLogger(__name__)
+
+
+class CastOutputToFloat(torch.nn.Module):
+    def __init__(self, layer):
+        super().__init__()
+        self.layer = layer
+
+    def forward(self, *args, **kwargs):
+        return self.layer(*args, **kwargs).float()
+
+
+def main():
+    parser = HfArgumentParser((ModelArguments, DataTrainingArguments, Seq2SeqTrainingArguments))
+    if len(sys.argv) == 2 and sys.argv[1].endswith(".json"):
+        # If we pass only one argument to the script and it's the path to a json file,
+        # let's parse it to get our arguments.
+        model_args, data_args, training_args = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1]))
+    else:
+        model_args, data_args, training_args = parser.parse_args_into_dataclasses()
+    # Setup logging
+    logging.basicConfig(
+        format="%(asctime)s - %(levelname)s - %(name)s - %(message)s",
+        datefmt="%m/%d/%Y %H:%M:%S",
+        handlers=[logging.StreamHandler(sys.stdout)],
+    )
+
+    if training_args.should_log:
+        # The default of training_args.log_level is passive, so we set log level at info here to have that default.
+        transformers.utils.logging.set_verbosity_info()
+
+    log_level = training_args.get_process_log_level()
+    logger.setLevel(log_level)
+    # datasets.utils.logging.set_verbosity(log_level)
+    transformers.utils.logging.set_verbosity(log_level)
+    transformers.utils.logging.enable_default_handler()
+    transformers.utils.logging.enable_explicit_format()
+
+    # Log on each process the small summary:
+    logger.warning(
+        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}"
+        + f"distributed training: {bool(training_args.local_rank != -1)}, 16-bits training: {training_args.fp16}"
+    )
+    logger.info(f"Training/evaluation parameters {training_args}")
+
+    # Set seed before initializing model.
+    set_seed(training_args.seed)
+
+    # Load pretrained model and tokenizer
+    tokenizer = AutoTokenizer.from_pretrained(model_args.model_name_or_path, trust_remote_code=True)
+    model = AutoModel.from_pretrained(model_args.model_name_or_path, trust_remote_code=True).half().cuda()
+
+    if model_args.quantization_bit is not None:
+        print(f"Quantized to {model_args.quantization_bit} bit")
+        model = model.quantize(model_args.quantization_bit)
+
+    with open(data_args.train_file, "r", encoding="utf-8") as f:
+        if data_args.train_file.endswith(".json"):
+            train_data = json.load(f)
+        elif data_args.train_file.endswith(".jsonl"):
+            train_data = [json.loads(line) for line in f]
+
+    if data_args.train_format == "input-output":
+        train_dataset = InputOutputDataset(
+            train_data,
+            tokenizer,
+            data_args.max_source_length,
+            data_args.max_target_length,
+        )
+    else:
+        raise ValueError(f"Unknown train format: {data_args.train_format}")
+    print(f"Train dataset size: {len(train_dataset)}")
+    #if training_args.local_rank < 1:
+    sanity_check(train_dataset[0]['input_ids'], train_dataset[0]['labels'], tokenizer)
+
+    # Apply PEFT configuration
+    peft_config = LoraConfig(
+        task_type=TaskType.CAUSAL_LM,
+        inference_mode=False,
+        r=model_args.lora_rank,
+        target_modules=['query_key_value'],
+        lora_alpha=model_args.lora_alpha,
+        lora_dropout=model_args.lora_dropout,
+    )
+    model = get_peft_model(model, peft_config).to("cuda")
+
+    # 确保梯度检查点和模型并行化设置正确
+    #model.gradient_checkpointing_enable()
+    model.enable_input_require_grads()
+    model.is_parallelizable = True
+    model.model_parallel = True  # 可以尝试暂时关闭模型并行化来看是否解决问题
+    model.lm_head = CastOutputToFloat(model.transformer.output_layer)
+    model.config.use_cache = False
+
+    # Data collator
+    data_collator = DataCollatorForSeq2Seq(
+        tokenizer,
+        model=model,
+        label_pad_token_id=-100,
+        pad_to_multiple_of=None,
+        padding=False
+    )
+
+    # Initialize our Trainer
+    trainer = LoRATrainer(
+        model=model,
+        args=training_args,
+        train_dataset=train_dataset,
+        tokenizer=tokenizer,
+        data_collator=data_collator,
+    )
+    checkpoint = None
+    if training_args.resume_from_checkpoint is not None:
+        checkpoint = training_args.resume_from_checkpoint
+    # model.gradient_checkpointing_enable()
+    model.enable_input_require_grads()
+    trainer.train()
+    trainer.save_model()  # Saves the tokenizer too for easy upload
+    trainer.save_state()
+
+
+if __name__ == "__main__":
+    main()
\ No newline at end of file
diff --git a/finetune_basemodel_demo/inference.py b/finetune_basemodel_demo/inference.py
new file mode 100644
index 0000000000000000000000000000000000000000..3ab0cdfa1760407a7d37985dfcc72207c3834d76
--- /dev/null
+++ b/finetune_basemodel_demo/inference.py
@@ -0,0 +1,47 @@
+import argparse
+import torch
+from transformers import AutoConfig, AutoModel, AutoTokenizer
+import os
+from peft import get_peft_model, LoraConfig, TaskType
+
+# Argument Parser Setup
+parser = argparse.ArgumentParser()
+parser.add_argument("--model", type=str, default=None,
+                    help="The directory of the model")
+parser.add_argument("--tokenizer", type=str, default=None, help="Tokenizer path")
+parser.add_argument("--lora-path", type=str, default=None,
+                    help="Path to the LoRA model checkpoint")
+parser.add_argument("--device", type=str, default="cuda", help="Device to use for computation")
+parser.add_argument("--max-new-tokens", type=int, default=128, help="Maximum new tokens for generation")
+parser.add_argument("--lora-alpha", type=float, default=32, help="LoRA alpha")
+parser.add_argument("--lora-rank", type=int, default=8, help="LoRA r")
+parser.add_argument("--lora-dropout", type=float, default=0.1, help="LoRA dropout")
+
+args = parser.parse_args()
+
+if args.tokenizer is None:
+    args.tokenizer = args.model
+
+# Model and Tokenizer Configuration
+tokenizer = AutoTokenizer.from_pretrained(args.tokenizer, trust_remote_code=True)
+model = AutoModel.from_pretrained(args.model, load_in_8bit=False, trust_remote_code=True, device_map="auto").to(
+    args.device)
+
+# LoRA Model Configuration
+peft_config = LoraConfig(
+    task_type=TaskType.CAUSAL_LM, inference_mode=True,
+    target_modules=['query_key_value'],
+    r=args.lora_rank, lora_alpha=args.lora_alpha, lora_dropout=args.lora_dropout
+)
+model = get_peft_model(model, peft_config)
+if os.path.exists(args.lora_path):
+    model.load_state_dict(torch.load(args.lora_path), strict=False)
+
+# Interactive Prompt
+while True:
+    prompt = input("Prompt: ")
+    inputs = tokenizer(prompt, return_tensors="pt").to(args.device)
+    response = model.generate(input_ids=inputs["input_ids"],
+                              max_length=inputs["input_ids"].shape[-1] + args.max_new_tokens)
+    response = response[0, inputs["input_ids"].shape[-1]:]
+    print("Response:", tokenizer.decode(response, skip_special_tokens=True))
diff --git a/finetune_basemodel_demo/preprocess_utils.py b/finetune_basemodel_demo/preprocess_utils.py
new file mode 100644
index 0000000000000000000000000000000000000000..2a8f80ffbdaee1fa6c9f0787f145f99c0e136c3e
--- /dev/null
+++ b/finetune_basemodel_demo/preprocess_utils.py
@@ -0,0 +1,52 @@
+from transformers import PreTrainedTokenizer
+from torch.utils.data import Dataset
+from typing import Dict, List
+
+
+def sanity_check(tokens: List[int], target: List[int], tokenizer: PreTrainedTokenizer):
+    print("Sanity Check >>>>>>>>>>>>>")
+    for t, m in zip(tokens, target):
+        decoded =  tokenizer.tokenizer.index_special_tokens[t] \
+            if t in tokenizer.tokenizer.index_special_tokens \
+            else tokenizer.decode([t])
+        print("%20s: %6d -> %6d" % (repr(decoded), t, m))
+    print("<<<<<<<<<<<<< Sanity Check")
+
+    assert len(tokens) == len(target), f"length mismatch: {len(tokens)} vs {len(target)}"
+
+
+class InputOutputDataset(Dataset):
+    def __init__(self, data: List[dict], tokenizer: PreTrainedTokenizer, max_source_length: int, max_target_length: int):
+        super(InputOutputDataset, self).__init__()
+        self.tokenizer = tokenizer
+        self.max_source_length = max_source_length
+        self.max_target_length = max_target_length
+        self.max_seq_length = max_source_length + max_target_length + 1
+        self.data = data
+
+    def __len__(self):
+        return len(self.data)
+    
+    def __getitem__(self, i) -> dict:
+        data_item = self.data[i]
+
+        a_ids = self.tokenizer.encode(text=data_item['context'], add_special_tokens=True, truncation=True,
+                                         max_length=self.max_source_length)
+        b_ids = self.tokenizer.encode(text=data_item['target'], add_special_tokens=False, truncation=True,
+                                    max_length=self.max_target_length)
+
+        context_length = len(a_ids)
+        input_ids = a_ids + b_ids + [self.tokenizer.eos_token_id]
+        labels = [self.tokenizer.pad_token_id] * context_length + b_ids + [self.tokenizer.eos_token_id]
+        
+        pad_len = self.max_seq_length - len(input_ids)
+        input_ids = input_ids + [self.tokenizer.pad_token_id] * pad_len
+        labels = labels + [self.tokenizer.pad_token_id] * pad_len
+        labels = [(l if l != self.tokenizer.pad_token_id else -100) for l in labels]
+
+        assert len(input_ids) == len(labels), f"length mismatch: {len(input_ids)} vs {len(labels)}"
+
+        return {
+            "input_ids": input_ids,
+            "labels": labels
+        }
diff --git a/finetune_basemodel_demo/requirements.txt b/finetune_basemodel_demo/requirements.txt
new file mode 100644
index 0000000000000000000000000000000000000000..9353ebcdf704836aafe02d0f60812d0c9743bfbe
--- /dev/null
+++ b/finetune_basemodel_demo/requirements.txt
@@ -0,0 +1,7 @@
+tqdm
+datasets
+fsspec
+astunparse
+peft
+accelerate
+sentencepiece
\ No newline at end of file
diff --git a/finetune_basemodel_demo/scripts/finetune_lora.sh b/finetune_basemodel_demo/scripts/finetune_lora.sh
new file mode 100644
index 0000000000000000000000000000000000000000..991cc23d540f9d9e0618a2ca4a2160f035a135e3
--- /dev/null
+++ b/finetune_basemodel_demo/scripts/finetune_lora.sh
@@ -0,0 +1,43 @@
+#! /usr/bin/env bash
+
+set -ex
+
+LR=1e-4
+NUM_GPUS=4
+LORA_RANK=8
+LORA_ALPHA=32
+LORA_DROUPOUT=0.1
+
+MAX_SOURCE_LEN=512
+MAX_TARGET_LEN=128
+DEV_BATCH_SIZE=1
+GRAD_ACCUMULARION_STEPS=1
+MAX_STEP=500
+SAVE_INTERVAL=50
+MAX_SEQ_LEN=512
+
+RUN_NAME=text
+BASE_MODEL_PATH=THUDM/chatglm3-6b-base
+DATASET_PATH=data/alpaca_data.jsonl
+DATESTR=`date +%Y%m%d-%H%M%S`
+OUTPUT_DIR=output/${RUN_NAME}-${DATESTR}-${LR}
+MASTER_PORT=$(shuf -n 1 -i 10000-65535)
+
+mkdir -p $OUTPUT_DIR
+
+torchrun --standalone --nnodes=1 --nproc_per_node=$NUM_GPUS finetune.py \
+    --train_format input-output \
+    --train_file $DATASET_PATH \
+    --lora_rank $LORA_RANK \
+    --lora_alpha $LORA_ALPHA \
+    --lora_dropout $LORA_DROUPOUT \
+    --max_seq_length $MAX_SEQ_LEN \
+    --preprocessing_num_workers 1 \
+    --model_name_or_path $BASE_MODEL_PATH \
+    --output_dir $OUTPUT_DIR \
+    --per_device_train_batch_size $DEV_BATCH_SIZE \
+    --gradient_accumulation_steps $GRAD_ACCUMULARION_STEPS \
+    --max_steps $MAX_STEP \
+    --logging_steps 1 \
+    --save_steps $SAVE_INTERVAL \
+    --learning_rate $LR  2>&1 | tee ${OUTPUT_DIR}/train.log
diff --git a/finetune_basemodel_demo/scripts/formate_alpaca2jsonl.py b/finetune_basemodel_demo/scripts/formate_alpaca2jsonl.py
new file mode 100644
index 0000000000000000000000000000000000000000..dd924ec41d479a5a4241a0adea46924f6ec06052
--- /dev/null
+++ b/finetune_basemodel_demo/scripts/formate_alpaca2jsonl.py
@@ -0,0 +1,28 @@
+import argparse
+import json
+import tqdm
+
+def format_example(example: dict) -> dict:
+    context = f"Instruction: {example['instruction']}\n"
+    if example.get("input"):
+        context += f"Input: {example['input']}\n"
+    context += "Answer: "
+    target = example["output"]
+    return {"context": context, "target": target}
+
+def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--data_path", type=str, default="data/alpaca_data.json")
+    parser.add_argument("--save_path", type=str, default="data/alpaca_data.jsonl")
+
+    args = parser.parse_args()
+    print("args:", args)
+    with open(args.data_path) as f:
+        examples = json.load(f)
+
+    with open(args.save_path, 'w') as f:
+        for example in tqdm.tqdm(examples, desc="formatting.."):
+            f.write(json.dumps(format_example(example), ensure_ascii=False) + '\n')
+
+if __name__ == "__main__":
+    main()
diff --git a/finetune_basemodel_demo/trainer.py b/finetune_basemodel_demo/trainer.py
new file mode 100644
index 0000000000000000000000000000000000000000..299d2a6ba76c209942f721997430bdff99d793eb
--- /dev/null
+++ b/finetune_basemodel_demo/trainer.py
@@ -0,0 +1,52 @@
+# coding=utf-8
+# Copyright 2020-present the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""
+The Trainer class, to easily train a 🤗 Transformers from scratch or finetune it on a new task.
+"""
+import os
+from typing import Optional
+from transformers import Trainer
+
+import torch
+from transformers.modeling_utils import PreTrainedModel, unwrap_model
+from transformers.utils import logging
+
+logger = logging.get_logger(__name__)
+
+WEIGHTS_NAME = "pytorch_model.pt"
+TRAINING_ARGS_NAME = "training_args.bin"
+
+class LoRATrainer(Trainer):
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args, **kwargs)
+
+    def compute_loss(self, model, inputs, return_outputs=False):
+        return model(**inputs).loss
+
+    def save_model(self, output_dir=None, _internal_call=False):
+        output_dir = output_dir if output_dir is not None else self.args.output_dir
+        os.makedirs(output_dir, exist_ok=True)
+        logger.info(f"Saving model checkpoint to {output_dir}")
+
+        model_to_save = unwrap_model(self.model)
+        saved_params = {
+            k: v.to("cuda") for k, v in model_to_save.named_parameters() if v.requires_grad
+        }
+        torch.save(saved_params, os.path.join(output_dir, WEIGHTS_NAME))
+
+        if self.tokenizer is not None:
+            self.tokenizer.save_pretrained(output_dir)
+
+        torch.save(self.args, os.path.join(output_dir, TRAINING_ARGS_NAME))
diff --git a/finetune_chatmodel_demo/README.md b/finetune_chatmodel_demo/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..990f2c678519ebd9ee9eb05f515ef896ab636f11
--- /dev/null
+++ b/finetune_chatmodel_demo/README.md
@@ -0,0 +1,261 @@
+# ChatGLM3-6B 微调示例
+
+本目录提供 ChatGLM3-6B 模型的微调示例，包括全量微调和 P-Tuning v2。格式上，提供多轮对话微调样例和输入输出格式微调样例。
+
+如果将模型下载到了本地，本文和代码中的 `THUDM/chatglm3-6b` 字段均应替换为相应地址以从本地加载模型。
+
+运行示例需要 `python>=3.10`，除基础的 `torch` 依赖外，示例代码运行还需要依赖 
+
+```bash
+pip install requirements.txt
+```
+
+## 多轮对话格式
+
+多轮对话微调示例采用 ChatGLM3 对话格式约定，对不同角色添加不同 `loss_mask` 从而在一遍计算中为多轮回复计算 `loss`。
+
+### 数据格式和预处理
+
+对于数据文件，样例采用如下格式
+
+如果您仅希望微调模型的对话能力，而非工具能力，您应该按照以下格式整理数据。
+
+```json
+[
+  {
+    "conversations": [
+      {
+        "role": "system",
+        "content": "<system prompt text>"
+      },
+      {
+        "role": "user",
+        "content": "<user prompt text>"
+      },
+      {
+        "role": "assistant",
+        "content": "<assistant response text>"
+      }, 
+       // ... Muti Turn
+      {
+        "role": "user",
+        "content": "<user prompt text>"
+      },
+      {
+        "role": "assistant",
+        "content": "<assistant response text>"
+      }
+    ]
+  }
+  // ...
+]
+```
+
+**请注意，这种方法在微调的step较多的情况下会影响到模型的工具调用功能**
+
+如果您希望微调模型的对话和工具能力，您应该按照以下格式整理数据。
+
+```json
+[
+   {
+      "tools": [
+         // available tools, format is not restricted
+      ],
+      "conversations": [
+         {
+            "role": "system",
+            "content": "<system prompt text>"
+         },
+         {
+            "role": "user",
+            "content": "<user prompt text>"
+         },
+         {
+            "role": "assistant",
+            "content": "<assistant thought to text>"
+         },
+         {
+            "role": "tool",
+            "name": "<name of the tool to be called",
+            "parameters": {
+               "<parameter_name>": "<parameter_value>"
+            },
+            "observation": "<observation>"
+            // don't have to be string
+         },
+         {
+            "role": "assistant",
+            "content": "<assistant response to observation>"
+         },
+         // ... Muti Turn
+         {
+            "role": "user",
+            "content": "<user prompt text>"
+         },
+         {
+            "role": "assistant",
+            "content": "<assistant response text>"
+         }
+      ]
+   }
+   // ...
+]
+```
+
+- 关于工具描述的 system prompt 无需手动插入，预处理时会将 `tools` 字段使用 `json.dumps(..., ensure_ascii=False)` 格式化后插入为首条 system prompt。
+
+- 每种角色可以附带一个 `bool` 类型的 `loss` 字段，表示该字段所预测的内容是否参与 `loss` 计算。若没有该字段，样例实现中默认对 `system`, `user` 不计算 `loss`，其余角色则计算 `loss`。
+
+- `tool` 并不是 ChatGLM3 中的原生角色，这里的 `tool` 在预处理阶段将被自动转化为一个具有工具调用 `metadata` 的 `assistant` 角色（默认计算 `loss`）和一个表示工具返回值的 `observation` 角色（不计算 `loss`）。
+
+- 目前暂未实现 `Code interpreter`的微调任务。
+
+- `system` 角色为可选角色，但若存在 `system` 角色，其必须出现在 `user` 角色之前，且一个完整的对话数据（无论单轮或者多轮对话）只能出现一次 `system` 角色。
+
+作为示例，我们使用 ToolAlpaca 数据集来进行微调。首先，克隆 [ToolAlpaca 数据集](https://github.com/tangqiaoyu/ToolAlpaca)，并使用
+
+```bash
+./scripts/format_tool_alpaca.py --path "ToolAlpaca/data/train_data.json"
+```
+
+将数据集处理成上述格式。在这里，我们有意将工具处理成了了 `list[str]` 这样的自然语言形式，以观察模型在微调前后对工具定义的理解能力。
+
+### 微调模型
+
+以下脚本提供了微调模型的参考方式。
+
+```bash
+./scripts/finetune_ds_multiturn.sh  # 全量微调
+./scripts/finetune_pt_multiturn.sh  # P-Tuning v2 微调
+```
+
+### 部署
+
+我们更新了 ChatGLM3 的综合 Demo，使其可以部署微调后的模型 checkpoint。
+
+对于全量微调，可以使用以下方式进行部署
+
+```bash
+cd ../composite_demo
+MODEL_PATH="path to finetuned model checkpoint" TOKENIZER_PATH="THUDM/chatglm3-6b" streamlit run main.py
+```
+
+对于 P-Tuning v2 微调，可以使用以下方式进行部署
+
+```bash
+cd ../composite_demo
+MODEL_PATH="THUDM/chatglm3-6b" PT_PATH="path to p-tuning checkpoint" streamlit run main.py
+```
+
+## 输入输出格式
+
+对于输入-输出格式，样例采用如下输入格式
+
+```json
+[
+  {
+    "prompt": "<prompt text>",
+    "response": "<response text>"
+  }
+  // ...
+]
+```
+
+预处理时，不会拼接任何角色标识符。
+
+作为示例，我们使用 AdvertiseGen 数据集来进行微调。从 [Google Drive](https://drive.google.com/file/d/13_vf0xRTQsyneRKdD1bZIr93vBGOczrk/view?usp=sharing) 或者 [Tsinghua Cloud](https://cloud.tsinghua.edu.cn/f/b3f119a008264b1cabd1/?dl=1) 下载处理好的 AdvertiseGen 数据集，将解压后的 `AdvertiseGen` 目录放到本目录下。
+
+```bash
+./scripts/format_advertise_gen.py --path "AdvertiseGen/train.json"
+```
+
+来下载和将数据集处理成上述格式。
+
+### 微调模型
+
+以下脚本提供了微调模型的参考方式。
+
+```bash
+./scripts/finetune_ds.sh  # 全量微调
+./scripts/finetune_pt.sh  # P-Tuning v2 微调
+```
+
+### 推理验证
+
+对于输入输出格式的微调，可使用 `inference.py` 进行基本的推理验证。
+
+```bash
+python inference.py \
+    --pt-checkpoint "path to p-tuning checkpoint" \
+    --model THUDM/chatglm3-6b 
+```
+
+```bash
+python inference.py \
+    --tokenizer THUDM/chatglm3-6b \
+    --model "path to finetuned model checkpoint" 
+```
+
+### 提示
+
+1. 微调代码在开始训练前，会先打印首条训练数据的预处理信息，显示为
+
+    ```log
+    Sanity Check >>>>>>>>>>>>>
+             '[gMASK]':  64790 ->   -100
+                 'sop':  64792 ->   -100
+          '<|system|>':  64794 ->   -100
+                    '':  30910 ->   -100
+                  '\n':     13 ->   -100
+              'Answer':  20115 ->   -100
+                 'the':    267 ->   -100
+           'following':   1762 ->   -100
+                      ...
+                'know':    683 ->   -100
+                 'the':    267 ->   -100
+            'response':   3010 ->   -100
+             'details':   3296 ->   -100
+                   '.':  30930 ->   -100
+       '<|assistant|>':  64796 ->   -100
+                    '':  30910 ->  30910
+                  '\n':     13 ->     13
+                   'I':    307 ->    307
+                'need':    720 ->    720
+                  'to':    289 ->    289
+                 'use':    792 ->    792
+                      ...
+    <<<<<<<<<<<<< Sanity Check
+    ```
+
+    字样，每行依次表示一个 detokenized string, token_id 和 target_id。可在日志中查看这部分的 `loss_mask` 是否符合预期。若不符合，可能需要调整代码或数据。
+
+2. 参考显存用量
+
+    - P-Tuning V2 `PRE_SEQ_LEN=128`, `DEV_BATCH_SIZE=1`, `GRAD_ACCUMULARION_STEPS=16`,  `MAX_SEQ_LEN=2048` 配置下约需要 21GB 显存。
+    - 全量微调时，`./scripts/finetune_ds_multiturn.sh` 中的配置（`MAX_SEQ_LEN=2048`, `DEV_BATCH_SIZE=16`, `GRAD_ACCUMULARION_STEPS=1`）恰好用满 4 * 80GB 显存。
+
+3. 若尝试后发现显存不足，可以考虑
+    - 尝试降低 `DEV_BATCH_SIZE` 并提升 `GRAD_ACCUMULARION_STEPS`
+    - 尝试添加 `--quantization_bit 8` 或 `--quantization_bit 4`。
+        - `PRE_SEQ_LEN=128`, `DEV_BATCH_SIZE=1`, `GRAD_ACCUMULARION_STEPS=16`, `MAX_SEQ_LEN=1024` 配置下，`--quantization_bit 8` 约需 12GB 显存，`--quantization_bit 4` 约需 7.6GB 显存。
+
+## 参考文献
+
+```
+@inproceedings{liu2022p,
+  title={P-tuning: Prompt tuning can be comparable to fine-tuning across scales and tasks},
+  author={Liu, Xiao and Ji, Kaixuan and Fu, Yicheng and Tam, Weng and Du, Zhengxiao and Yang, Zhilin and Tang, Jie},
+  booktitle={Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)},
+  pages={61--68},
+  year={2022}
+}
+
+@misc{tang2023toolalpaca,
+      title={ToolAlpaca: Generalized Tool Learning for Language Models with 3000 Simulated Cases}, 
+      author={Qiaoyu Tang and Ziliang Deng and Hongyu Lin and Xianpei Han and Qiao Liang and Le Sun},
+      year={2023},
+      eprint={2306.05301},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+```
diff --git a/finetune_chatmodel_demo/arguments.py b/finetune_chatmodel_demo/arguments.py
new file mode 100644
index 0000000000000000000000000000000000000000..4ed9c65b1bc82ec1c0d5809885e7c1d3a4df8f7e
--- /dev/null
+++ b/finetune_chatmodel_demo/arguments.py
@@ -0,0 +1,150 @@
+from dataclasses import dataclass, field
+from typing import Optional
+
+
+@dataclass
+class ModelArguments:
+    """
+    Arguments pertaining to which model/config/tokenizer we are going to fine-tune from.
+    """
+
+    model_name_or_path: str = field(
+        metadata={"help": "Path to pretrained model or model identifier from huggingface.co/models"}
+    )
+    ptuning_checkpoint: str = field(
+        default=None, metadata={"help": "Path to p-tuning v2 checkpoints"}
+    )
+    config_name: Optional[str] = field(
+        default=None, metadata={"help": "Pretrained config name or path if not the same as model_name"}
+    )
+    tokenizer_name: Optional[str] = field(
+        default=None, metadata={"help": "Pretrained tokenizer name or path if not the same as model_name"}
+    )
+    cache_dir: Optional[str] = field(
+        default=None,
+        metadata={"help": "Where to store the pretrained models downloaded from huggingface.co"},
+    )
+    use_fast_tokenizer: bool = field(
+        default=True,
+        metadata={"help": "Whether to use one of the fast tokenizer (backed by the tokenizers library) or not."},
+    )
+    model_revision: str = field(
+        default="main",
+        metadata={"help": "The specific model version to use (can be a branch name, tag name or commit id)."},
+    )
+    use_auth_token: bool = field(
+        default=False,
+        metadata={
+            "help": (
+                "Will use the token generated when running `huggingface-cli login` (necessary to use this script "
+                "with private models)."
+            )
+        },
+    )
+    resize_position_embeddings: Optional[bool] = field(
+        default=None,
+        metadata={
+            "help": (
+                "Whether to automatically resize the position embeddings if `max_source_length` exceeds "
+                "the model's position embeddings."
+            )
+        },
+    )
+    quantization_bit: Optional[int] = field(
+        default=None
+    )
+    pre_seq_len: Optional[int] = field(
+        default=None
+    )
+    prefix_projection: bool = field(
+        default=False
+    )
+
+
+@dataclass
+class DataTrainingArguments:
+    """
+    Arguments pertaining to what data we are going to input our model for training and eval.
+    """
+    train_file: Optional[str] = field(
+        default=None, metadata={"help": "The input training data file (a jsonlines or csv file)."}
+    )
+
+    max_seq_length: Optional[int] = field(
+        default=2048,
+        metadata={
+            "help": (
+                "The maximum total input sequence length after tokenization. Sequences longer "
+                "than this will be truncated."
+            )
+        },
+    )
+
+    max_source_length: Optional[int] = field(
+        default=1024,
+        metadata={
+            "help": (
+                "The maximum total input sequence length after tokenization. Sequences longer "
+                "than this will be truncated, sequences shorter will be padded."
+            )
+        },
+    )
+    max_target_length: Optional[int] = field(
+        default=128,
+        metadata={
+            "help": (
+                "The maximum total sequence length for target text after tokenization. Sequences longer "
+                "than this will be truncated, sequences shorter will be padded."
+            )
+        },
+    )
+
+    train_format: str = field(
+        default=None, metadata={"help": "The format of the training data file (mulit-turn or input-output)"},
+    )
+
+    overwrite_cache: bool = field(
+        default=False, metadata={"help": "Overwrite the cached training and evaluation sets"}
+    )
+
+    preprocessing_num_workers: Optional[int] = field(
+        default=None,
+        metadata={"help": "The number of processes to use for the preprocessing."},
+    )
+
+    max_seq_length: Optional[int] = field(
+        default=1024,
+        metadata={
+            "help": (
+                "The maximum total input sequence length after tokenization. Sequences longer "
+                "than this will be truncated, sequences shorter will be padded."
+            )
+        },
+    )
+
+    pad_to_max_length: bool = field(
+        default=False,
+        metadata={
+            "help": (
+                "Whether to pad all samples to model maximum sentence length. "
+                "If False, will pad the samples dynamically when batching to the maximum length in the batch. More "
+                "efficient on GPU but very bad for TPU."
+            )
+        },
+    )
+
+    max_train_samples: Optional[int] = field(
+        default=None,
+        metadata={
+            "help": (
+                "For debugging purposes or quicker training, truncate the number of training examples to this "
+                "value if set."
+            )
+        },
+    )
+    
+    def __post_init__(self):
+        extension = self.train_file.split(".")[-1]
+        assert extension in {"jsonl", "json"}, "`train_file` should be a jsonl or a json file."
+
+        assert self.train_format in {"multi-turn", "input-output"}
\ No newline at end of file
diff --git a/finetune_chatmodel_demo/configs/deepspeed.json b/finetune_chatmodel_demo/configs/deepspeed.json
new file mode 100644
index 0000000000000000000000000000000000000000..798932966f38b2df8a468c72a4b41d8b47033ccc
--- /dev/null
+++ b/finetune_chatmodel_demo/configs/deepspeed.json
@@ -0,0 +1,21 @@
+{
+  "train_micro_batch_size_per_gpu": "auto",
+  "zero_allow_untested_optimizer": true,
+  "fp16": {
+    "enabled": "auto",
+    "loss_scale": 0,
+    "initial_scale_power": 16,
+    "loss_scale_window": 1000,
+    "hysteresis": 2,
+    "min_loss_scale": 1
+  },
+  "zero_optimization": {
+    "stage": 2,
+    "allgather_partitions": true,
+    "allgather_bucket_size": 5e8,
+    "overlap_comm": false,
+    "reduce_scatter": true,
+    "reduce_bucket_size": 5e8,
+    "contiguous_gradients" : true
+  }
+}
\ No newline at end of file
diff --git a/finetune_chatmodel_demo/finetune.py b/finetune_chatmodel_demo/finetune.py
new file mode 100644
index 0000000000000000000000000000000000000000..a2b7c46e539c7745538c7d426cd42de34a51d34f
--- /dev/null
+++ b/finetune_chatmodel_demo/finetune.py
@@ -0,0 +1,165 @@
+#!/usr/bin/env python
+# coding=utf-8
+# Copyright 2021 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""
+Fine-tuning the library models for sequence to sequence.
+"""
+# You can also adapt this script on your own sequence to sequence task. Pointers for this are left as comments.
+# Adapted from 
+
+
+import logging
+import os
+import sys
+import torch
+import json
+import transformers
+from transformers import (
+    AutoConfig,
+    AutoModel,
+    AutoTokenizer,
+    DataCollatorForSeq2Seq,
+    HfArgumentParser,
+    Seq2SeqTrainingArguments,
+    set_seed,
+)
+from trainer import PrefixTrainer
+
+from arguments import ModelArguments, DataTrainingArguments
+
+from preprocess_utils import sanity_check, MultiTurnDataset, InputOutputDataset
+
+logger = logging.getLogger(__name__)
+
+def main():
+    parser = HfArgumentParser((ModelArguments, DataTrainingArguments, Seq2SeqTrainingArguments))
+    if len(sys.argv) == 2 and sys.argv[1].endswith(".json"):
+        # If we pass only one argument to the script and it's the path to a json file,
+        # let's parse it to get our arguments.
+        model_args, data_args, training_args = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1]))
+    else:
+        model_args, data_args, training_args = parser.parse_args_into_dataclasses()
+
+    # Setup logging
+    logging.basicConfig(
+        format="%(asctime)s - %(levelname)s - %(name)s - %(message)s",
+        datefmt="%m/%d/%Y %H:%M:%S",
+        handlers=[logging.StreamHandler(sys.stdout)],
+    )
+
+    if training_args.should_log:
+        # The default of training_args.log_level is passive, so we set log level at info here to have that default.
+        transformers.utils.logging.set_verbosity_info()
+
+    log_level = training_args.get_process_log_level()
+    logger.setLevel(log_level)
+    # datasets.utils.logging.set_verbosity(log_level)
+    transformers.utils.logging.set_verbosity(log_level)
+    transformers.utils.logging.enable_default_handler()
+    transformers.utils.logging.enable_explicit_format()
+
+    # Log on each process the small summary:
+    logger.warning(
+        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}"
+        + f"distributed training: {bool(training_args.local_rank != -1)}, 16-bits training: {training_args.fp16}"
+    )
+    logger.info(f"Training/evaluation parameters {training_args}")
+
+    # Set seed before initializing model.
+    set_seed(training_args.seed)
+
+    # Load pretrained model and tokenizer
+    config = AutoConfig.from_pretrained(model_args.model_name_or_path, trust_remote_code=True)
+    config.pre_seq_len = model_args.pre_seq_len
+    config.prefix_projection = model_args.prefix_projection
+
+    tokenizer = AutoTokenizer.from_pretrained(model_args.model_name_or_path, trust_remote_code=True)
+
+    if model_args.ptuning_checkpoint is not None:
+        model = AutoModel.from_pretrained(model_args.model_name_or_path, config=config, trust_remote_code=True)
+        prefix_state_dict = torch.load(os.path.join(model_args.ptuning_checkpoint, "pytorch_model.bin"))
+        new_prefix_state_dict = {}
+        for k, v in prefix_state_dict.items():
+            if k.startswith("transformer.prefix_encoder."):
+                new_prefix_state_dict[k[len("transformer.prefix_encoder."):]] = v
+        model.transformer.prefix_encoder.load_state_dict(new_prefix_state_dict)
+    else:
+        model = AutoModel.from_pretrained(model_args.model_name_or_path, config=config, trust_remote_code=True)
+
+    if model_args.quantization_bit is not None:
+        print(f"Quantized to {model_args.quantization_bit} bit")
+        model = model.quantize(model_args.quantization_bit)
+    if model_args.pre_seq_len is not None:
+        # P-tuning v2
+        model = model.half()
+        model.transformer.prefix_encoder.float()
+    else:
+        # Finetune
+        model = model.float()
+    
+    with open(data_args.train_file, "r", encoding="utf-8") as f:
+        if data_args.train_file.endswith(".json"):
+            train_data = json.load(f)
+        elif data_args.train_file.endswith(".jsonl"):
+            train_data = [json.loads(line) for line in f]
+
+    if data_args.train_format == "multi-turn":
+        train_dataset = MultiTurnDataset(
+            train_data,
+            tokenizer,
+            data_args.max_seq_length,
+        )
+    elif data_args.train_format == "input-output":
+        train_dataset = InputOutputDataset(
+            train_data,
+            tokenizer,
+            data_args.max_source_length,
+            data_args.max_target_length,
+        )
+    else:
+        raise ValueError(f"Unknown train format: {data_args.train_format}")
+    if training_args.local_rank < 1:
+        sanity_check(train_dataset[0]['input_ids'], train_dataset[0]['labels'], tokenizer)
+
+    # Data collator
+    data_collator = DataCollatorForSeq2Seq(
+        tokenizer,
+        model=model,
+        label_pad_token_id=-100,
+        pad_to_multiple_of=None,
+        padding=False
+    )
+
+    # Initialize our Trainer
+    trainer = PrefixTrainer(
+        model=model,
+        args=training_args,
+        train_dataset=train_dataset,
+        tokenizer=tokenizer,
+        data_collator=data_collator,
+        save_changed=model_args.pre_seq_len is not None
+    )
+
+    checkpoint = None
+    if training_args.resume_from_checkpoint is not None:
+        checkpoint = training_args.resume_from_checkpoint
+    model.gradient_checkpointing_enable()
+    model.enable_input_require_grads()
+    trainer.train(resume_from_checkpoint=checkpoint)
+    trainer.save_model()  # Saves the tokenizer too for easy upload
+    trainer.save_state()
+
+if __name__ == "__main__":
+    main()
diff --git a/finetune_chatmodel_demo/inference.py b/finetune_chatmodel_demo/inference.py
new file mode 100644
index 0000000000000000000000000000000000000000..b25e91ab75dea4258e747c59688f6f15af46e054
--- /dev/null
+++ b/finetune_chatmodel_demo/inference.py
@@ -0,0 +1,41 @@
+import argparse
+from transformers import AutoConfig, AutoModel, AutoTokenizer
+import torch
+import os
+
+parser = argparse.ArgumentParser()
+parser.add_argument("--pt-checkpoint", type=str, default=None, help="The checkpoint path")
+parser.add_argument("--model", type=str, default=None, help="main model weights")
+parser.add_argument("--tokenizer", type=str, default=None, help="main model weights")
+parser.add_argument("--pt-pre-seq-len", type=int, default=128, help="The pre-seq-len used in p-tuning")
+parser.add_argument("--device", type=str, default="cuda")
+parser.add_argument("--max-new-tokens", type=int, default=128)
+
+args = parser.parse_args()
+
+if args.tokenizer is None:
+    args.tokenizer = args.model
+
+if args.pt_checkpoint:
+    tokenizer = AutoTokenizer.from_pretrained(args.tokenizer, trust_remote_code=True)
+    config = AutoConfig.from_pretrained(args.model, trust_remote_code=True, pre_seq_len=args.pt_pre_seq_len)
+    model = AutoModel.from_pretrained(args.model, config=config, trust_remote_code=True).cuda()
+    prefix_state_dict = torch.load(os.path.join(args.pt_checkpoint, "pytorch_model.bin"))
+    new_prefix_state_dict = {}
+    for k, v in prefix_state_dict.items():
+        if k.startswith("transformer.prefix_encoder."):
+            new_prefix_state_dict[k[len("transformer.prefix_encoder."):]] = v
+    model.transformer.prefix_encoder.load_state_dict(new_prefix_state_dict)
+else:
+    tokenizer = AutoTokenizer.from_pretrained(args.tokenizer, trust_remote_code=True)
+    model = AutoModel.from_pretrained(args.model, trust_remote_code=True)
+
+model = model.to(args.device)
+
+while True:
+    prompt = input("Prompt:")
+    inputs = tokenizer(prompt, return_tensors="pt")
+    inputs = inputs.to(args.device)
+    response = model.generate(input_ids=inputs["input_ids"], max_length=inputs["input_ids"].shape[-1] + args.max_new_tokens)
+    response = response[0, inputs["input_ids"].shape[-1]:]
+    print("Response:", tokenizer.decode(response, skip_special_tokens=True))
\ No newline at end of file
diff --git a/finetune_chatmodel_demo/preprocess_utils.py b/finetune_chatmodel_demo/preprocess_utils.py
new file mode 100644
index 0000000000000000000000000000000000000000..2cc7bc2de5cfe3c4738f52308c28d0f3d25d643b
--- /dev/null
+++ b/finetune_chatmodel_demo/preprocess_utils.py
@@ -0,0 +1,146 @@
+import json
+import ast
+import astunparse
+from transformers import PreTrainedTokenizer
+from torch.utils.data import Dataset
+from copy import deepcopy
+from typing import Dict, List
+
+# text constants
+FUNCTION_CALL_NAME     = 'tool_call'
+FUNCTION_CALL_PREFIX   = '```python\n'
+FUNCTION_CALL_POSTFIX  = '\n```'
+TOOL_DEFINITION_PREFIX = 'Answer the following questions as best as you can. You have access to the following tools:\n'
+CONVERSATOIN_KEY       = 'conversations'
+TOOL_DESC_KEY          = 'tools'
+
+def format_function_call(function_name: str, parameters: Dict[str, str]):
+    function_name = ast.Name(id=function_name)
+    keywords = [
+        ast.keyword(arg=arg_name, value=ast.Constant(arg_value)) 
+        for arg_name, arg_value in parameters.items()
+    ]
+    func_call = ast.Call(func=function_name, args=[], keywords=keywords)
+    return astunparse.unparse(func_call).strip()
+
+def format_conversation(item, tokenizer, conversation_key: str, tool_key: str):
+    conversations = deepcopy(item[conversation_key])
+
+    # Note: `loss_mask` here means whether *the prediction* of the token should take loss
+    tokens, loss_masks = [tokenizer.get_command("[gMASK]"), tokenizer.get_command("sop")], [0, 0]
+
+    def _update(_tokens: List[int], value: int = 1):
+        value = int(value)
+        tokens.extend(_tokens)
+        loss_masks.extend([value] * len(_tokens))
+
+    # insert system prompt for tools
+    if tool_key in item:
+        conversations.insert(0, 
+            {
+                "role": "system", 
+                "content": TOOL_DEFINITION_PREFIX + json.dumps(item[tool_key], indent=4, ensure_ascii=False)
+            }
+        )
+    
+    for idx, conv in enumerate(conversations):
+        loss = conv.get("loss", True)
+        if conv['role'] in {'system', 'user'}:
+            loss = False
+        if conv['role'] == 'tool':
+            # function call python code
+            value = FUNCTION_CALL_PREFIX + format_function_call(FUNCTION_CALL_NAME, conv["parameters"]) + FUNCTION_CALL_POSTFIX
+            text = tokenizer.build_single_message("assistant", conv["name"], value)
+            _update(text, loss)
+
+            # function call result
+            value = conv.get('observation', None)
+            if not isinstance(value, str):
+                value = json.dumps(value, ensure_ascii=False)
+            text = tokenizer.build_single_message("observation", "", value)
+            _update(text, False)
+        else:
+            text = tokenizer.build_single_message(conv['role'], "", conv["content"])
+            _update(text, loss)
+
+    _update([tokenizer.eos_token_id], False)
+
+    assert len(tokens) == len(loss_masks), f"length mismatch: {len(tokens)} vs {len(loss_masks)}"
+    return tokens, loss_masks
+
+def sanity_check(tokens: List[int], target: List[int], tokenizer: PreTrainedTokenizer):
+    print("Sanity Check >>>>>>>>>>>>>")
+    for t, m in zip(tokens, target):
+        decoded =  tokenizer.tokenizer.index_special_tokens[t] \
+            if t in tokenizer.tokenizer.index_special_tokens \
+            else tokenizer.decode([t])
+        print("%20s: %6d -> %6d" % (repr(decoded), t, m))
+    print("<<<<<<<<<<<<< Sanity Check")
+
+    assert len(tokens) == len(target), f"length mismatch: {len(tokens)} vs {len(target)}"
+
+class MultiTurnDataset(Dataset):
+    def __init__(self, data: List[dict], tokenizer: PreTrainedTokenizer, max_seq_length: int):
+        super(MultiTurnDataset, self).__init__()
+        self.tokenizer = tokenizer
+        self.max_seq_length = max_seq_length
+        self.data = data
+
+    def __len__(self):
+        return len(self.data)
+
+    def __getitem__(self, i) -> dict:
+        data_item = self.data[i]
+        tokens, loss_masks = format_conversation(data_item, self.tokenizer, CONVERSATOIN_KEY, TOOL_DESC_KEY)
+
+        # labels are used inside the model
+        target_based_loss_mask = [False] + loss_masks[:-1]
+        labels = [(t if m else -100) for t, m in zip(tokens, target_based_loss_mask)]
+
+        tokens = tokens[:self.max_seq_length]
+        labels = labels[:self.max_seq_length]
+        tokens += [self.tokenizer.pad_token_id] * (self.max_seq_length - len(tokens))
+        labels += [-100] * (self.max_seq_length - len(labels))
+
+        assert len(tokens) == len(labels), f"length mismatch: {len(tokens)} vs {len(labels)}"
+
+        return {
+            "input_ids": tokens,
+            "labels": labels
+        }
+    
+class InputOutputDataset(Dataset):
+    def __init__(self, data: List[dict], tokenizer: PreTrainedTokenizer, max_source_length: int, max_target_length: int):
+        super(InputOutputDataset, self).__init__()
+        self.tokenizer = tokenizer
+        self.max_source_length = max_source_length
+        self.max_target_length = max_target_length
+        self.max_seq_length = max_source_length + max_target_length + 1
+        self.data = data
+
+    def __len__(self):
+        return len(self.data)
+    
+    def __getitem__(self, i) -> dict:
+        data_item = self.data[i]
+
+        a_ids = self.tokenizer.encode(text=data_item['prompt'], add_special_tokens=True, truncation=True,
+                                         max_length=self.max_source_length)
+        b_ids = self.tokenizer.encode(text=data_item['response'], add_special_tokens=False, truncation=True,
+                                    max_length=self.max_target_length)
+
+        context_length = len(a_ids)
+        input_ids = a_ids + b_ids + [self.tokenizer.eos_token_id]
+        labels = [self.tokenizer.pad_token_id] * context_length + b_ids + [self.tokenizer.eos_token_id]
+        
+        pad_len = self.max_seq_length - len(input_ids)
+        input_ids = input_ids + [self.tokenizer.pad_token_id] * pad_len
+        labels = labels + [self.tokenizer.pad_token_id] * pad_len
+        labels = [(l if l != self.tokenizer.pad_token_id else -100) for l in labels]
+
+        assert len(input_ids) == len(labels), f"length mismatch: {len(input_ids)} vs {len(labels)}"
+
+        return {
+            "input_ids": input_ids,
+            "labels": labels
+        }
diff --git a/finetune_chatmodel_demo/requirements.txt b/finetune_chatmodel_demo/requirements.txt
new file mode 100644
index 0000000000000000000000000000000000000000..774714277548dea05a301fe5daf6033f45aded7a
--- /dev/null
+++ b/finetune_chatmodel_demo/requirements.txt
@@ -0,0 +1,5 @@
+transformers==4.30.2
+accelerate
+sentencepiece
+astunparse
+deepspeed
\ No newline at end of file
diff --git a/finetune_chatmodel_demo/scripts/finetune_ds.sh b/finetune_chatmodel_demo/scripts/finetune_ds.sh
new file mode 100644
index 0000000000000000000000000000000000000000..0fc4e7dccac599181c65b340c1bf1b928e0a37b5
--- /dev/null
+++ b/finetune_chatmodel_demo/scripts/finetune_ds.sh
@@ -0,0 +1,39 @@
+#! /usr/bin/env bash
+
+set -ex
+
+LR=1e-4
+NUM_GPUS=4
+MAX_SOURCE_LEN=1024
+MAX_TARGET_LEN=128
+DEV_BATCH_SIZE=4
+GRAD_ACCUMULARION_STEPS=1
+MAX_STEP=500
+SAVE_INTERVAL=500
+
+RUN_NAME=advertise_gen_ft
+BASE_MODEL_PATH=THUDM/chatglm3-6b
+DATASET_PATH=formatted_data/advertise_gen.jsonl
+
+DATESTR=`date +%Y%m%d-%H%M%S`
+OUTPUT_DIR=output/${RUN_NAME}-${DATESTR}-${LR}
+MASTER_PORT=$(shuf -n 1 -i 10000-65535)
+
+mkdir -p $OUTPUT_DIR
+
+torchrun --standalone --nnodes=1 --nproc_per_node=$NUM_GPUS finetune.py \
+    --train_format input-output \
+    --train_file $DATASET_PATH \
+    --preprocessing_num_workers 1 \
+    --model_name_or_path $BASE_MODEL_PATH \
+    --output_dir $OUTPUT_DIR \
+    --max_source_length $MAX_SOURCE_LEN \
+    --max_target_length $MAX_TARGET_LEN \
+    --per_device_train_batch_size $DEV_BATCH_SIZE \
+    --gradient_accumulation_steps $GRAD_ACCUMULARION_STEPS \
+    --max_steps $MAX_STEP \
+    --logging_steps 1 \
+    --save_steps $SAVE_INTERVAL \
+    --learning_rate $LR \
+    --fp16 \
+    --deepspeed configs/deepspeed.json 2>&1 | tee ${OUTPUT_DIR}/train.log
diff --git a/finetune_chatmodel_demo/scripts/finetune_ds_multiturn.sh b/finetune_chatmodel_demo/scripts/finetune_ds_multiturn.sh
new file mode 100644
index 0000000000000000000000000000000000000000..ff91e9301290808d39540946e52ce77bd39ee889
--- /dev/null
+++ b/finetune_chatmodel_demo/scripts/finetune_ds_multiturn.sh
@@ -0,0 +1,35 @@
+#! /usr/bin/env bash
+
+set -ex
+
+LR=1e-4
+NUM_GPUS=4
+MAX_SEQ_LEN=2048
+DEV_BATCH_SIZE=16
+GRAD_ACCUMULARION_STEPS=1
+MAX_STEP=200
+SAVE_INTERVAL=50
+
+DATESTR=`date +%Y%m%d-%H%M%S`
+RUN_NAME=tool_alpaca_ft
+DATASET_PATH=formatted_data/tool_alpaca.jsonl
+
+BASE_MODEL_PATH=THUDM/chatglm3-6b
+OUTPUT_DIR=output/${RUN_NAME}-${DATESTR}-${LR}
+
+mkdir -p $OUTPUT_DIR
+
+torchrun --standalone --nnodes=1 --nproc_per_node=$NUM_GPUS finetune.py \
+    --train_format multi-turn \
+    --train_file $DATASET_PATH \
+    --max_seq_length $MAX_SEQ_LEN \
+    --preprocessing_num_workers 1 \
+    --model_name_or_path $BASE_MODEL_PATH \
+    --output_dir $OUTPUT_DIR \
+    --per_device_train_batch_size $DEV_BATCH_SIZE \
+    --gradient_accumulation_steps $GRAD_ACCUMULARION_STEPS \
+    --max_steps $MAX_STEP \
+    --logging_steps 1 \
+    --save_steps $SAVE_INTERVAL \
+    --fp16 \
+    --deepspeed configs/deepspeed.json 2>&1 | tee ${OUTPUT_DIR}/train.log
diff --git a/finetune_chatmodel_demo/scripts/finetune_pt.sh b/finetune_chatmodel_demo/scripts/finetune_pt.sh
new file mode 100644
index 0000000000000000000000000000000000000000..42ac8496a0ecf16275c468ee2d6106df593cc72f
--- /dev/null
+++ b/finetune_chatmodel_demo/scripts/finetune_pt.sh
@@ -0,0 +1,38 @@
+#! /usr/bin/env bash
+
+set -ex
+
+PRE_SEQ_LEN=128
+LR=2e-2
+NUM_GPUS=1
+MAX_SOURCE_LEN=1024
+MAX_TARGET_LEN=128
+DEV_BATCH_SIZE=1
+GRAD_ACCUMULARION_STEPS=32
+MAX_STEP=1000
+SAVE_INTERVAL=500
+
+DATESTR=`date +%Y%m%d-%H%M%S`
+RUN_NAME=advertise_gen_pt
+
+BASE_MODEL_PATH=THUDM/chatglm3-6b
+DATASET_PATH=formatted_data/advertise_gen.jsonl
+OUTPUT_DIR=output/${RUN_NAME}-${DATESTR}-${PRE_SEQ_LEN}-${LR}
+
+mkdir -p $OUTPUT_DIR
+
+torchrun --standalone --nnodes=1 --nproc_per_node=$NUM_GPUS finetune.py \
+    --train_format input-output \
+    --train_file $DATASET_PATH \
+    --preprocessing_num_workers 1 \
+    --model_name_or_path $BASE_MODEL_PATH \
+    --output_dir $OUTPUT_DIR \
+    --max_source_length $MAX_SOURCE_LEN \
+    --max_target_length $MAX_TARGET_LEN \
+    --per_device_train_batch_size $DEV_BATCH_SIZE \
+    --gradient_accumulation_steps $GRAD_ACCUMULARION_STEPS \
+    --max_steps $MAX_STEP \
+    --logging_steps 1 \
+    --save_steps $SAVE_INTERVAL \
+    --learning_rate $LR \
+    --pre_seq_len $PRE_SEQ_LEN 2>&1 | tee ${OUTPUT_DIR}/train.log
diff --git a/finetune_chatmodel_demo/scripts/finetune_pt_multiturn.sh b/finetune_chatmodel_demo/scripts/finetune_pt_multiturn.sh
new file mode 100644
index 0000000000000000000000000000000000000000..e021d07ed815e82fd7860ae16accf97f980976f0
--- /dev/null
+++ b/finetune_chatmodel_demo/scripts/finetune_pt_multiturn.sh
@@ -0,0 +1,36 @@
+#! /usr/bin/env bash
+
+set -ex
+
+PRE_SEQ_LEN=128
+LR=2e-2
+NUM_GPUS=1
+MAX_SEQ_LEN=2048
+DEV_BATCH_SIZE=1
+GRAD_ACCUMULARION_STEPS=16
+MAX_STEP=1000
+SAVE_INTERVAL=500
+
+DATESTR=`date +%Y%m%d-%H%M%S`
+RUN_NAME=tool_alpaca_pt
+
+BASE_MODEL_PATH=THUDM/chatglm3-6b
+DATASET_PATH=formatted_data/tool_alpaca.jsonl
+OUTPUT_DIR=output/${RUN_NAME}-${DATESTR}-${PRE_SEQ_LEN}-${LR}
+
+mkdir -p $OUTPUT_DIR
+
+torchrun --standalone --nnodes=1 --nproc_per_node=$NUM_GPUS finetune.py \
+    --train_format multi-turn \
+    --train_file $DATASET_PATH \
+    --max_seq_length $MAX_SEQ_LEN \
+    --preprocessing_num_workers 1 \
+    --model_name_or_path $BASE_MODEL_PATH \
+    --output_dir $OUTPUT_DIR \
+    --per_device_train_batch_size $DEV_BATCH_SIZE \
+    --gradient_accumulation_steps $GRAD_ACCUMULARION_STEPS \
+    --max_steps $MAX_STEP \
+    --logging_steps 1 \
+    --save_steps $SAVE_INTERVAL \
+    --learning_rate $LR \
+    --pre_seq_len $PRE_SEQ_LEN 2>&1 | tee ${OUTPUT_DIR}/train.log
diff --git a/finetune_chatmodel_demo/scripts/format_advertise_gen.py b/finetune_chatmodel_demo/scripts/format_advertise_gen.py
new file mode 100644
index 0000000000000000000000000000000000000000..b3c71d43fdf83f0bab918c7a6291130fc2418f9e
--- /dev/null
+++ b/finetune_chatmodel_demo/scripts/format_advertise_gen.py
@@ -0,0 +1,25 @@
+#! /usr/bin/env python
+
+import json
+from collections import Counter
+from argparse import ArgumentParser
+import os
+
+parser = ArgumentParser()
+parser.add_argument("--path", type=str, required=True)
+
+args = parser.parse_args()
+
+with open(args.path) as f:
+    data = [json.loads(line) for line in f]
+
+train_examples = [{
+    "prompt": x['content'],
+    "response": x['summary'],
+} for x in data]
+
+os.makedirs("formatted_data", exist_ok=True)
+
+with open("formatted_data/advertise_gen.jsonl", "w") as f:
+    for e in train_examples:
+        f.write(json.dumps(e, ensure_ascii=False) + "\n")
diff --git a/finetune_chatmodel_demo/scripts/format_tool_alpaca.py b/finetune_chatmodel_demo/scripts/format_tool_alpaca.py
new file mode 100644
index 0000000000000000000000000000000000000000..a9efbfe147246c45f566211be62cee3a0743ca5c
--- /dev/null
+++ b/finetune_chatmodel_demo/scripts/format_tool_alpaca.py
@@ -0,0 +1,60 @@
+#! /usr/bin/env python
+
+import json
+from collections import Counter
+from argparse import ArgumentParser
+import os
+
+parser = ArgumentParser()
+parser.add_argument("--path", type=str, required=True)
+
+args = parser.parse_args()
+
+with open(args.path) as f:
+    data = json.load(f)
+
+train_examples = []
+err_count = 0
+for setting in data:
+    api_desc = [setting["NLDocumentation"]]
+    for instance in setting["Instances"]:
+        try:
+            conv = [{
+                "role": "user",
+                "content": instance['input'],
+            }]
+            for step in instance['intermediate_steps']:
+                tool_name, params, react = step[0]
+                step_thought = react.split("Action:")[0].strip()
+                observation = step[1]
+                conv.append({
+                    "role": "assistant",
+                    "content": step_thought,
+                })
+                conv.append({
+                    "role": "tool",
+                    "name": tool_name,
+                    "parameters": json.loads(params),
+                    "observation": observation,
+                })
+            conv.append({
+                "role": "assistant",
+                "content": instance['Final Thought'] + "\n" + instance['output'],
+            })
+        except:
+            err_count += 1
+        else:
+            train_examples.append({
+                "tools": api_desc,
+                "conversations": conv
+            })
+
+print("err_count:", err_count)
+print("train_examples:", len(train_examples))
+print("conversation distribution:", Counter([len(e["conversations"]) for e in train_examples]))
+
+os.makedirs("formatted_data", exist_ok=True)
+
+with open("formatted_data/tool_alpaca.jsonl", "w") as f:
+    for e in train_examples:
+        f.write(json.dumps(e, ensure_ascii=False) + "\n")
\ No newline at end of file
diff --git a/finetune_chatmodel_demo/trainer.py b/finetune_chatmodel_demo/trainer.py
new file mode 100644
index 0000000000000000000000000000000000000000..0f8366d12beeb03e996d453baaae19812f5dec54
--- /dev/null
+++ b/finetune_chatmodel_demo/trainer.py
@@ -0,0 +1,72 @@
+# coding=utf-8
+# Copyright 2020-present the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""
+The Trainer class, to easily train a 🤗 Transformers from scratch or finetune it on a new task.
+"""
+import os
+from typing import Optional
+from transformers import Trainer
+
+import torch
+from transformers.modeling_utils import PreTrainedModel, unwrap_model
+from transformers.utils import logging
+
+logger = logging.get_logger(__name__)
+
+WEIGHTS_NAME = "pytorch_model.bin"
+TRAINING_ARGS_NAME = "training_args.bin"
+
+
+class PrefixTrainer(Trainer):
+    def __init__(self, *args, save_changed=False, **kwargs):
+        self.save_changed = save_changed
+        super().__init__(*args, **kwargs)
+
+    def _save(self, output_dir: Optional[str] = None, state_dict=None):
+        # If we are executing this function, we are the process zero, so we don't check for that.
+        output_dir = output_dir if output_dir is not None else self.args.output_dir
+        os.makedirs(output_dir, exist_ok=True)
+        logger.info(f"Saving model checkpoint to {output_dir}")
+        # Save a trained model and configuration using `save_pretrained()`.
+        # They can then be reloaded using `from_pretrained()`
+        if not isinstance(self.model, PreTrainedModel):
+            if isinstance(unwrap_model(self.model), PreTrainedModel):
+                if state_dict is None:
+                    state_dict = self.model.state_dict()
+                unwrap_model(self.model).save_pretrained(output_dir, state_dict=state_dict)
+            else:
+                logger.info("Trainer.model is not a `PreTrainedModel`, only saving its state dict.")
+                if state_dict is None:
+                    state_dict = self.model.state_dict()
+                torch.save(state_dict, os.path.join(output_dir, WEIGHTS_NAME))
+        else:
+            if self.save_changed:
+                print("Saving PrefixEncoder")
+                state_dict = self.model.state_dict()
+                filtered_state_dict = {}
+                for k, v in self.model.named_parameters():
+                    if v.requires_grad:
+                        filtered_state_dict[k] = state_dict[k]
+                self.model.save_pretrained(output_dir, state_dict=filtered_state_dict)
+            else:
+                print("Saving the whole model")
+                self.model.save_pretrained(output_dir, state_dict=state_dict)
+        if self.tokenizer is not None:
+            self.tokenizer.save_pretrained(output_dir)
+
+        # Good practice: save your training arguments together with the trained model
+        torch.save(self.args, os.path.join(output_dir, TRAINING_ARGS_NAME))
+
+#TODO: add LoRATrainer Support Later
diff --git a/langchain_demo/ChatGLM3.py b/langchain_demo/ChatGLM3.py
new file mode 100644
index 0000000000000000000000000000000000000000..6190932b59a89f973f34ad962fe774a749c7d309
--- /dev/null
+++ b/langchain_demo/ChatGLM3.py
@@ -0,0 +1,123 @@
+import json
+from langchain.llms.base import LLM
+from transformers import AutoTokenizer, AutoModel, AutoConfig
+from typing import List, Optional
+from utils import tool_config_from_file
+
+
+class ChatGLM3(LLM):
+    max_token: int = 8192
+    do_sample: bool = False
+    temperature: float = 0.8
+    top_p = 0.8
+    tokenizer: object = None
+    model: object = None
+    history: List = []
+    tool_names: List = []
+    has_search: bool = False
+
+    def __init__(self):
+        super().__init__()
+
+    @property
+    def _llm_type(self) -> str:
+        return "ChatGLM3"
+
+    def load_model(self, model_name_or_path=None):
+        model_config = AutoConfig.from_pretrained(
+            model_name_or_path,
+            trust_remote_code=True
+        )
+        self.tokenizer = AutoTokenizer.from_pretrained(
+            model_name_or_path,
+            trust_remote_code=True
+        )
+        self.model = AutoModel.from_pretrained(
+            model_name_or_path, config=model_config, trust_remote_code=True
+        ).half().cuda()
+
+    def _tool_history(self, prompt: str):
+        ans = []
+        tool_prompts = prompt.split(
+            "You have access to the following tools:\n\n")[1].split("\n\nUse a json blob")[0].split("\n")
+
+        tool_names = [tool.split(":")[0] for tool in tool_prompts]
+        self.tool_names = tool_names
+        tools_json = []
+        for i, tool in enumerate(tool_names):
+            tool_config = tool_config_from_file(tool)
+            if tool_config:
+                tools_json.append(tool_config)
+            else:
+                ValueError(
+                    f"Tool {tool} config not found! It's description is {tool_prompts[i]}"
+                )
+
+        ans.append({
+            "role": "system",
+            "content": "Answer the following questions as best as you can. You have access to the following tools:",
+            "tools": tools_json
+        })
+        query = f"""{prompt.split("Human: ")[-1].strip()}"""
+        return ans, query
+
+    def _extract_observation(self, prompt: str):
+        return_json = prompt.split("Observation: ")[-1].split("\nThought:")[0]
+        self.history.append({
+            "role": "observation",
+            "content": return_json
+        })
+        return
+
+    def _extract_tool(self):
+        if len(self.history[-1]["metadata"]) > 0:
+            metadata = self.history[-1]["metadata"]
+            content = self.history[-1]["content"]
+            if "tool_call" in content:
+                for tool in self.tool_names:
+                    if tool in metadata:
+                        input_para = content.split("='")[-1].split("'")[0]
+                        action_json = {
+                            "action": tool,
+                            "action_input": input_para
+                        }
+                        self.has_search = True
+                        return f"""
+Action: 
+```
+{json.dumps(action_json, ensure_ascii=False)}
+```"""
+        final_answer_json = {
+            "action": "Final Answer",
+            "action_input": self.history[-1]["content"]
+        }
+        self.has_search = False
+        return f"""
+Action: 
+```
+{json.dumps(final_answer_json, ensure_ascii=False)}
+```"""
+
+    def _call(self, prompt: str, history: List = [], stop: Optional[List[str]] = ["<|user|>"]):
+        print("======")
+        print(prompt)
+        print("======")
+        if not self.has_search:
+            self.history, query = self._tool_history(prompt)
+        else:
+            self._extract_observation(prompt)
+            query = ""
+        # print("======")
+        # print(history)
+        # print("======")
+        _, self.history = self.model.chat(
+            self.tokenizer,
+            query,
+            history=self.history,
+            do_sample=self.do_sample,
+            max_length=self.max_token,
+            temperature=self.temperature,
+        )
+        response = self._extract_tool()
+        history.append((prompt, response))
+        return response
diff --git a/langchain_demo/README.md b/langchain_demo/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..a71763345e1d6b3c5632720fea608a9a2973a362
--- /dev/null
+++ b/langchain_demo/README.md
@@ -0,0 +1,71 @@
+# README
+
+## 模型配置
+
+在 `main.py` 文件中，修改 `model_path = /path/to/chatglm3-6b` 路径，也可以填写 `THUDM/chatglm3-6b` 自动下载模型。
+
+## 工具添加
+
+### LangChain 已实现工具
+
+参考 [langchain](https://python.langchain.com/docs/modules/agents/tools/) 工具相关函数，在 `main.py` 中导入工具模块，例如导入 `arxiv` 工具
+
+```python
+run_tool(["arxiv"], llm, [
+    "帮我查询AgentTuning相关工作"
+])
+```
+
+#### Calculator、Weather Tool配置 
+
+如果你的 Python 环境中 `LangChain` 的版本低于  **`0.0.278`**  则需要在这两个预定义工具类中实现 `_arun` 方法
+否则将会出现 
+`TypeError: Can't instantiate abstract class Weather with abstract method _arun`
+
+示例如下：
+```python
+class Weather(BaseTool):
+    name = "weather"
+    description = "Use for searching weather at a specific location"
+
+    async def _arun(self, *args: Any, **kwargs: Any) -> Any:
+        # 用例中没有用到 arun 不予具体实现
+        pass
+```
+
+运行 `main.py` 文件
+
+```
+python main.py
+```
+
+模型会因找不到 `arxiv` 工具的 yaml 文件描述而中断，需要用户手动构建 `./Tool/arxiv.yaml` 文件。工具可以用户自行描述，也可以参考 LangChain 对该工具的描述。
+
+对 `arxiv` 这个例子而言，参考内容位于 `./Tool/arxiv_example.yaml` 文件，可参考该文件构建 `Tool/arxiv.yaml` 文件（最简单的方式修改名称即可），重新运行模型就能合理调用工具。
+
+> 有些工具需要导入 API_KEY，按照 langchain 报错添加到环境变量即可。
+
+### 自定义工具
+
+如果用户想自定义工具，可以参考 `Tool/Weather.py` 以及 `Tool/Weather.yaml` 文件，重载新的 `Tool` 类，实现其对应的 `_run()` 方法，然后在 `main.py` 中导入该工具模块，例如导入 `Weather` 工具，即可以调用
+
+```python
+# 对同一个工具调用多次
+# 需要 export SENIVERSE_KEY=<YOUR_API_KEY_HERE>
+run_tool([Weather()], llm, [
+        "今天北京天气怎么样？",
+        "What's the weather like in Shanghai today",
+])
+```
+
+## 多工具使用
+
+可以将多个工具组装在一起让模型自动选择调用，例如
+
+```python
+run_tool([Calculator(), "arxiv", Weather()], llm, [
+        "帮我检索GLM-130B相关论文",
+        "今天北京天气怎么样？",
+        "根号3减去根号二再加上4等于多少？",
+])
+```
diff --git a/langchain_demo/Tool/Calculator.py b/langchain_demo/Tool/Calculator.py
new file mode 100644
index 0000000000000000000000000000000000000000..099115225362de9b971110b0caa27435fd1defa9
--- /dev/null
+++ b/langchain_demo/Tool/Calculator.py
@@ -0,0 +1,32 @@
+import abc
+import math
+from typing import Any
+
+from langchain.tools import BaseTool
+
+
+class Calculator(BaseTool, abc.ABC):
+    name = "Calculator"
+    description = "Useful for when you need to answer questions about math"
+
+    def __init__(self):
+        super().__init__()
+
+    async def _arun(self, *args: Any, **kwargs: Any) -> Any:
+        # 用例中没有用到 arun 不予具体实现
+        pass
+
+
+    def _run(self, para: str) -> str:
+        para = para.replace("^", "**")
+        if "sqrt" in para:
+            para = para.replace("sqrt", "math.sqrt")
+        elif "log" in para:
+            para = para.replace("log", "math.log")
+        return eval(para)
+
+
+if __name__ == "__main__":
+    calculator_tool = Calculator()
+    result = calculator_tool.run("sqrt(2) + 3")
+    print(result)
diff --git a/langchain_demo/Tool/Calculator.yaml b/langchain_demo/Tool/Calculator.yaml
new file mode 100644
index 0000000000000000000000000000000000000000..74c4a19ce2ea257fc454b214e0c0e34b0299715c
--- /dev/null
+++ b/langchain_demo/Tool/Calculator.yaml
@@ -0,0 +1,10 @@
+name: Calculator
+description: Useful for when you need to answer questions about math
+parameters:
+  type: object
+  properties:
+    formula:
+      type: string
+      description: The formula to be calculated
+  required:
+    - formula
\ No newline at end of file
diff --git a/langchain_demo/Tool/Weather.py b/langchain_demo/Tool/Weather.py
new file mode 100644
index 0000000000000000000000000000000000000000..6f41facfabdd7075ffb87aebe2f392c49c23b767
--- /dev/null
+++ b/langchain_demo/Tool/Weather.py
@@ -0,0 +1,41 @@
+import os
+from typing import Any
+
+import requests
+from langchain.tools import BaseTool
+
+
+class Weather(BaseTool):
+    name = "weather"
+    description = "Use for searching weather at a specific location"
+
+    def __init__(self):
+        super().__init__()
+
+    async def _arun(self, *args: Any, **kwargs: Any) -> Any:
+        # 用例中没有用到 arun 不予具体实现
+        pass
+
+    def get_weather(self, location):
+        api_key = os.environ["SENIVERSE_KEY"]
+        url = f"https://api.seniverse.com/v3/weather/now.json?key={api_key}&location={location}&language=zh-Hans&unit=c"
+        response = requests.get(url)
+        if response.status_code == 200:
+            data = response.json()
+            weather = {
+                "temperature": data["results"][0]["now"]["temperature"],
+                "description": data["results"][0]["now"]["text"],
+            }
+            return weather
+        else:
+            raise Exception(
+                f"Failed to retrieve weather: {response.status_code}")
+
+    def _run(self, para: str) -> str:
+        return self.get_weather(para)
+
+
+if __name__ == "__main__":
+    weather_tool = Weather()
+    weather_info = weather_tool.run("成都")
+    print(weather_info)
diff --git a/langchain_demo/Tool/arxiv_example.yaml b/langchain_demo/Tool/arxiv_example.yaml
new file mode 100644
index 0000000000000000000000000000000000000000..5da8a9f9543aa3cef4cde663df79c15014ccad57
--- /dev/null
+++ b/langchain_demo/Tool/arxiv_example.yaml
@@ -0,0 +1,10 @@
+name: arxiv
+description: A wrapper around Arxiv.org for searching and retrieving scientific articles in various fields.
+parameters:
+  type: object
+  properties:
+    query:
+      type: string
+      description: The search query title
+required:
+  - query
\ No newline at end of file
diff --git a/langchain_demo/Tool/weather.yaml b/langchain_demo/Tool/weather.yaml
new file mode 100644
index 0000000000000000000000000000000000000000..77bc4eccb2c3c130a6fb32fa175b432ae6ebeefb
--- /dev/null
+++ b/langchain_demo/Tool/weather.yaml
@@ -0,0 +1,10 @@
+name: weather
+description: Search the current weather of a city
+parameters:
+  type: object
+  properties:
+    city:
+      type: string
+      description: City name
+  required:
+    - city
\ No newline at end of file
diff --git a/langchain_demo/main.py b/langchain_demo/main.py
new file mode 100644
index 0000000000000000000000000000000000000000..d610c46290a6cd17d9b3e30a527360f80169768b
--- /dev/null
+++ b/langchain_demo/main.py
@@ -0,0 +1,57 @@
+import os
+from typing import List
+from ChatGLM3 import ChatGLM3
+
+from langchain.agents import load_tools
+from Tool.Weather import Weather
+from Tool.Calculator import Calculator
+from langchain.agents import initialize_agent
+from langchain.agents import AgentType
+
+MODEL_PATH = os.environ.get('MODEL_PATH', 'THUDM/chatglm3-6b')
+
+def run_tool(tools, llm, prompt_chain: List[str]):
+    loaded_tolls = []
+    for tool in tools:
+        if isinstance(tool, str):
+            loaded_tolls.append(load_tools([tool], llm=llm)[0])
+        else:
+            loaded_tolls.append(tool)
+    agent = initialize_agent(
+        loaded_tolls, llm,
+        agent=AgentType.STRUCTURED_CHAT_ZERO_SHOT_REACT_DESCRIPTION,
+        verbose=True,
+        handle_parsing_errors=True
+    )
+    for prompt in prompt_chain:
+        agent.run(prompt)
+
+
+if __name__ == "__main__":
+    llm = ChatGLM3()
+    llm.load_model(model_name_or_path=MODEL_PATH)
+
+    # arxiv: 单个工具调用示例 1
+    run_tool(["arxiv"], llm, [
+        "帮我查询GLM-130B相关工作"
+    ])
+
+    # weather: 单个工具调用示例 2
+    run_tool([Weather()], llm, [
+        "今天北京天气怎么样？",
+        "What's the weather like in Shanghai today",
+    ])
+
+    # calculator: 单个工具调用示例 3
+    run_tool([Calculator()], llm, [
+        "12345679乘以54等于多少？",
+        "3.14的3.14次方等于多少？",
+        "根号2加上根号三等于多少？",
+    ]),
+
+    # arxiv + weather + calculator: 多个工具结合调用
+    # run_tool([Calculator(), "arxiv", Weather()], llm, [
+    #     "帮我检索GLM-130B相关论文",
+    #     "今天北京天气怎么样？",
+    #     "根号3减去根号二再加上4等于多少？",
+    # ])
\ No newline at end of file
diff --git a/langchain_demo/requirements.txt b/langchain_demo/requirements.txt
new file mode 100644
index 0000000000000000000000000000000000000000..4cd8e9cc84b34a8c1fde1bedfa852ea4d95c7825
--- /dev/null
+++ b/langchain_demo/requirements.txt
@@ -0,0 +1,2 @@
+langchain
+arxiv
\ No newline at end of file
diff --git a/langchain_demo/utils.py b/langchain_demo/utils.py
new file mode 100644
index 0000000000000000000000000000000000000000..ec04d40cd02651b07d591bff33c6b10fa7fd616a
--- /dev/null
+++ b/langchain_demo/utils.py
@@ -0,0 +1,12 @@
+import os
+import yaml
+
+
+def tool_config_from_file(tool_name, directory="Tool/"):
+    """search tool yaml and return json format"""
+    for filename in os.listdir(directory):
+        if filename.endswith('.yaml') and tool_name in filename:
+            file_path = os.path.join(directory, filename)
+            with open(file_path, encoding='utf-8') as f:
+                return yaml.safe_load(f)
+    return None
diff --git a/openai_api_demo/openai_api.py b/openai_api_demo/openai_api.py
new file mode 100644
index 0000000000000000000000000000000000000000..6895657606c8dbab4f9b1177a55bcf88049bf3c1
--- /dev/null
+++ b/openai_api_demo/openai_api.py
@@ -0,0 +1,249 @@
+# coding=utf-8
+# Implements API for ChatGLM3-6B in OpenAI's format. (https://platform.openai.com/docs/api-reference/chat)
+# Usage: python openai_api.py
+# Visit http://localhost:8000/docs for documents.
+
+# 在OpenAI的API中，max_tokens 等价于 HuggingFace 的 max_new_tokens 而不是 max_length，。
+# 例如，对于6b模型，设置max_tokens = 8192，则会报错，因为扣除历史记录和提示词后，模型不能输出那么多的tokens。
+
+import os
+import time
+from contextlib import asynccontextmanager
+from typing import List, Literal, Optional, Union
+
+import torch
+import uvicorn
+from fastapi import FastAPI, HTTPException
+from fastapi.middleware.cors import CORSMiddleware
+from loguru import logger
+from pydantic import BaseModel, Field
+from sse_starlette.sse import EventSourceResponse
+from transformers import AutoTokenizer, AutoModel
+
+from utils import process_response, generate_chatglm3, generate_stream_chatglm3
+
+MODEL_PATH = os.environ.get('MODEL_PATH', 'THUDM/chatglm3-6b')
+TOKENIZER_PATH = os.environ.get("TOKENIZER_PATH", MODEL_PATH)
+DEVICE = 'cuda' if torch.cuda.is_available() else 'cpu'
+
+
+@asynccontextmanager
+async def lifespan(app: FastAPI):  # collects GPU memory
+    yield
+    if torch.cuda.is_available():
+        torch.cuda.empty_cache()
+        torch.cuda.ipc_collect()
+
+
+app = FastAPI(lifespan=lifespan)
+
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+
+class ModelCard(BaseModel):
+    id: str
+    object: str = "model"
+    created: int = Field(default_factory=lambda: int(time.time()))
+    owned_by: str = "owner"
+    root: Optional[str] = None
+    parent: Optional[str] = None
+    permission: Optional[list] = None
+
+
+class ModelList(BaseModel):
+    object: str = "list"
+    data: List[ModelCard] = []
+
+
+class FunctionCallResponse(BaseModel):
+    name: Optional[str] = None
+    arguments: Optional[str] = None
+
+
+class ChatMessage(BaseModel):
+    role: Literal["user", "assistant", "system", "function"]
+    content: str = None
+    name: Optional[str] = None
+    function_call: Optional[FunctionCallResponse] = None
+
+
+class DeltaMessage(BaseModel):
+    role: Optional[Literal["user", "assistant", "system"]] = None
+    content: Optional[str] = None
+    function_call: Optional[FunctionCallResponse] = None
+
+
+class ChatCompletionRequest(BaseModel):
+    model: str
+    messages: List[ChatMessage]
+    temperature: Optional[float] = 0.8
+    top_p: Optional[float] = 0.8
+    max_tokens: Optional[int] = None
+    stream: Optional[bool] = False
+    functions: Optional[Union[dict, List[dict]]] = None
+    # Additional parameters
+    repetition_penalty: Optional[float] = 1.1
+
+
+class ChatCompletionResponseChoice(BaseModel):
+    index: int
+    message: ChatMessage
+    finish_reason: Literal["stop", "length", "function_call"]
+
+
+class ChatCompletionResponseStreamChoice(BaseModel):
+    index: int
+    delta: DeltaMessage
+    finish_reason: Optional[Literal["stop", "length", "function_call"]]
+
+
+class UsageInfo(BaseModel):
+    prompt_tokens: int = 0
+    total_tokens: int = 0
+    completion_tokens: Optional[int] = 0
+
+
+class ChatCompletionResponse(BaseModel):
+    model: str
+    object: Literal["chat.completion", "chat.completion.chunk"]
+    choices: List[Union[ChatCompletionResponseChoice, ChatCompletionResponseStreamChoice]]
+    created: Optional[int] = Field(default_factory=lambda: int(time.time()))
+    usage: Optional[UsageInfo] = None
+
+
+@app.get("/v1/models", response_model=ModelList)
+async def list_models():
+    model_card = ModelCard(id="chatglm3-6b")
+    return ModelList(data=[model_card])
+
+
+@app.post("/v1/chat/completions", response_model=ChatCompletionResponse)
+async def create_chat_completion(request: ChatCompletionRequest):
+    global model, tokenizer
+
+    if len(request.messages) < 1 or request.messages[-1].role == "assistant":
+        raise HTTPException(status_code=400, detail="Invalid request")
+
+    gen_params = dict(
+        messages=request.messages,
+        temperature=request.temperature,
+        top_p=request.top_p,
+        max_tokens=request.max_tokens or 1024,
+        echo=False,
+        stream=request.stream,
+        repetition_penalty=request.repetition_penalty,
+        functions=request.functions,
+    )
+
+    logger.debug(f"==== request ====\n{gen_params}")
+
+    if request.stream:
+        generate = predict(request.model, gen_params)
+        return EventSourceResponse(generate, media_type="text/event-stream")
+
+    response = generate_chatglm3(model, tokenizer, gen_params)
+
+    # Remove the first newline character
+    if response["text"].startswith("\n"):
+        response["text"] = response["text"][1:]
+    response["text"] = response["text"].strip()
+    usage = UsageInfo()
+    function_call, finish_reason = None, "stop"
+    if request.functions:
+        try:
+            function_call = process_response(response["text"], use_tool=True)
+        except:
+            logger.warning("Failed to parse tool call, maybe the response is not a tool call or have been answered.")
+
+    if isinstance(function_call, dict):
+        finish_reason = "function_call"
+        function_call = FunctionCallResponse(**function_call)
+
+    message = ChatMessage(
+        role="assistant",
+        content=response["text"],
+        function_call=function_call if isinstance(function_call, FunctionCallResponse) else None,
+    )
+
+    logger.debug(f"==== message ====\n{message}")
+
+    choice_data = ChatCompletionResponseChoice(
+        index=0,
+        message=message,
+        finish_reason=finish_reason,
+    )
+    task_usage = UsageInfo.model_validate(response["usage"])
+    for usage_key, usage_value in task_usage.model_dump().items():
+        setattr(usage, usage_key, getattr(usage, usage_key) + usage_value)
+    return ChatCompletionResponse(model=request.model, choices=[choice_data], object="chat.completion", usage=usage)
+
+
+async def predict(model_id: str, params: dict):
+    global model, tokenizer
+
+    choice_data = ChatCompletionResponseStreamChoice(
+        index=0,
+        delta=DeltaMessage(role="assistant"),
+        finish_reason=None
+    )
+    chunk = ChatCompletionResponse(model=model_id, choices=[choice_data], object="chat.completion.chunk")
+    yield "{}".format(chunk.model_dump_json(exclude_unset=True))
+
+    previous_text = ""
+    for new_response in generate_stream_chatglm3(model, tokenizer, params):
+        decoded_unicode = new_response["text"]
+        delta_text = decoded_unicode[len(previous_text):]
+        previous_text = decoded_unicode
+
+        finish_reason = new_response["finish_reason"]
+        if len(delta_text) == 0 and finish_reason != "function_call":
+            continue
+
+        function_call = None
+        if finish_reason == "function_call":
+            try:
+                function_call = process_response(decoded_unicode, use_tool=True)
+            except:
+                logger.warning("Failed to parse tool call, maybe the response is not a tool call or have been answered.")
+
+        if isinstance(function_call, dict):
+            function_call = FunctionCallResponse(**function_call)
+
+        delta = DeltaMessage(
+            content=delta_text,
+            role="assistant",
+            function_call=function_call if isinstance(function_call, FunctionCallResponse) else None,
+        )
+
+        choice_data = ChatCompletionResponseStreamChoice(
+            index=0,
+            delta=delta,
+            finish_reason=finish_reason
+        )
+        chunk = ChatCompletionResponse(model=model_id, choices=[choice_data], object="chat.completion.chunk")
+        yield "{}".format(chunk.model_dump_json(exclude_unset=True))
+
+    choice_data = ChatCompletionResponseStreamChoice(
+        index=0,
+        delta=DeltaMessage(),
+        finish_reason="stop"
+    )
+    chunk = ChatCompletionResponse(model=model_id, choices=[choice_data], object="chat.completion.chunk")
+    yield "{}".format(chunk.model_dump_json(exclude_unset=True))
+    yield '[DONE]'
+
+
+if __name__ == "__main__":
+
+    tokenizer = AutoTokenizer.from_pretrained(TOKENIZER_PATH, trust_remote_code=True)
+    if 'cuda' in DEVICE:  # AMD, NVIDIA GPU can use Half Precision
+        model = AutoModel.from_pretrained(MODEL_PATH, trust_remote_code=True).to(DEVICE).eval()
+    else:  # CPU, Intel GPU and other GPU can use Float16 Precision Only
+        model = AutoModel.from_pretrained(MODEL_PATH, trust_remote_code=True).float().to(DEVICE).eval()
+    uvicorn.run(app, host='0.0.0.0', port=8000, workers=1)
diff --git a/openai_api_demo/openai_api_request.py b/openai_api_demo/openai_api_request.py
new file mode 100644
index 0000000000000000000000000000000000000000..eea7f6569ca7a578b9a0da4c838567981452fb58
--- /dev/null
+++ b/openai_api_demo/openai_api_request.py
@@ -0,0 +1,113 @@
+# 使用curl命令测试返回
+# curl -X POST "http://127.0.0.1:8000/v1/chat/completions" \
+# -H "Content-Type: application/json" \
+# -d "{\"model\": \"chatglm3-6b\", \"messages\": [{\"role\": \"system\", \"content\": \"You are ChatGLM3, a large language model trained by Zhipu.AI. Follow the user's instructions carefully. Respond using markdown.\"}, {\"role\": \"user\", \"content\": \"你好，给我讲一个故事，大概100字\"}], \"stream\": false, \"max_tokens\": 100, \"temperature\": 0.8, \"top_p\": 0.8}"
+
+# 使用Python代码测返回
+import requests
+import json
+
+base_url = "http://127.0.0.1:8000"
+
+
+def create_chat_completion(model, messages, functions, use_stream=False):
+    data = {
+        "functions": functions,  # 函数定义
+        "model": model,  # 模型名称
+        "messages": messages,  # 会话历史
+        "stream": use_stream,  # 是否流式响应
+        "max_tokens": 100,  # 最多生成字数
+        "temperature": 0.8,  # 温度
+        "top_p": 0.8,  # 采样概率
+    }
+
+    response = requests.post(f"{base_url}/v1/chat/completions", json=data, stream=use_stream)
+    if response.status_code == 200:
+        if use_stream:
+            # 处理流式响应
+            for line in response.iter_lines():
+                if line:
+                    decoded_line = line.decode('utf-8')[6:]
+                    try:
+                        response_json = json.loads(decoded_line)
+                        content = response_json.get("choices", [{}])[0].get("delta", {}).get("content", "")
+                        print(content)
+                    except:
+                        print("Special Token:", decoded_line)
+        else:
+            # 处理非流式响应
+            decoded_line = response.json()
+            content = decoded_line.get("choices", [{}])[0].get("message", "").get("content", "")
+            print(content)
+    else:
+        print("Error:", response.status_code)
+        return None
+
+
+def function_chat(use_stream=True):
+    functions = [
+        {
+            "name": "get_current_weather",
+            "description": "Get the current weather in a given location.",
+            "parameters": {
+                "type": "object",
+                "properties": {
+                    "location": {
+                        "type": "string",
+                        "description": "The city and state, e.g. Beijing",
+                    },
+                    "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]},
+                },
+                "required": ["location"],
+            },
+        }
+    ]
+    chat_messages = [
+        {
+            "role": "user",
+            "content": "波士顿天气如何？",
+        },
+        {
+            "role": "assistant",
+            "content": "get_current_weather\n ```python\ntool_call(location='Beijing', unit='celsius')\n```",
+            "function_call": {
+                "name": "get_current_weather",
+                "arguments": '{"location": "Beijing", "unit": "celsius"}',
+            },
+        },
+        {
+            "role": "function",
+            "name": "get_current_weather",
+            "content": '{"temperature": "12", "unit": "celsius", "description": "Sunny"}',
+        },
+        # ... 接下来这段是 assistant 的回复和用户的回复。
+        # {
+        #     "role": "assistant",
+        #     "content": "根据最新的天气预报，目前北京的天气情况是晴朗的，温度为12摄氏度。",
+        # },
+        # {
+        #     "role": "user",
+        #     "content": "谢谢",
+        # }
+    ]
+    create_chat_completion("chatglm3-6b", messages=chat_messages, functions=functions, use_stream=use_stream)
+
+
+def simple_chat(use_stream=True):
+    functions = None
+    chat_messages = [
+        {
+            "role": "system",
+            "content": "You are ChatGLM3, a large language model trained by Zhipu.AI. Follow the user's instructions carefully. Respond using markdown.",
+        },
+        {
+            "role": "user",
+            "content": "你好，给我讲一个故事，大概100字"
+        }
+    ]
+    create_chat_completion("chatglm3-6b", messages=chat_messages, functions=functions, use_stream=use_stream)
+
+
+if __name__ == "__main__":
+    function_chat(use_stream=False)
+    # simple_chat(use_stream=True)
diff --git a/openai_api_demo/requirements.txt b/openai_api_demo/requirements.txt
new file mode 100644
index 0000000000000000000000000000000000000000..49d9a2dacadbd028fb10deb51fbd87bf626e58af
--- /dev/null
+++ b/openai_api_demo/requirements.txt
@@ -0,0 +1,2 @@
+openai>=1.3.5
+pydantic>=2.5.1
\ No newline at end of file
diff --git a/openai_api_demo/utils.py b/openai_api_demo/utils.py
new file mode 100644
index 0000000000000000000000000000000000000000..ddd4166ebc9ea785c27177f44cbd7824789d4741
--- /dev/null
+++ b/openai_api_demo/utils.py
@@ -0,0 +1,243 @@
+import os
+import gc
+import json
+import torch
+from torch.nn import Module
+from transformers import PreTrainedModel, PreTrainedTokenizer
+from transformers import AutoModel
+from transformers.generation.logits_process import LogitsProcessor
+from typing import Dict, Union, Optional, Tuple
+
+
+def auto_configure_device_map(num_gpus: int) -> Dict[str, int]:
+    # transformer.word_embeddings 占用1层
+    # transformer.final_layernorm 和 lm_head 占用1层
+    # transformer.layers 占用 28 层
+    # 总共30层分配到num_gpus张卡上
+    num_trans_layers = 28
+    per_gpu_layers = 30 / num_gpus
+
+    # bugfix: 在linux中调用torch.embedding传入的weight,input不在同一device上,导致RuntimeError
+    # windows下 model.device 会被设置成 transformer.word_embeddings.device
+    # linux下 model.device 会被设置成 lm_head.device
+    # 在调用chat或者stream_chat时,input_ids会被放到model.device上
+    # 如果transformer.word_embeddings.device和model.device不同,则会导致RuntimeError
+    # 因此这里将transformer.word_embeddings,transformer.final_layernorm,lm_head都放到第一张卡上
+    # 本文件来源于https://github.com/THUDM/ChatGLM-6B/blob/main/utils.py
+    # 仅此处做少许修改以支持ChatGLM3
+    device_map = {
+        'transformer.embedding.word_embeddings': 0,
+        'transformer.encoder.final_layernorm': 0,
+        'transformer.output_layer': 0,
+        'transformer.rotary_pos_emb': 0,
+        'lm_head': 0
+    }
+
+    used = 2
+    gpu_target = 0
+    for i in range(num_trans_layers):
+        if used >= per_gpu_layers:
+            gpu_target += 1
+            used = 0
+        assert gpu_target < num_gpus
+        device_map[f'transformer.encoder.layers.{i}'] = gpu_target
+        used += 1
+
+    return device_map
+
+
+def load_model_on_gpus(checkpoint_path: Union[str, os.PathLike], num_gpus: int = 2,
+                       device_map: Optional[Dict[str, int]] = None, **kwargs) -> Module:
+    if num_gpus < 2 and device_map is None:
+        model = AutoModel.from_pretrained(checkpoint_path, trust_remote_code=True, **kwargs).half().cuda()
+    else:
+        from accelerate import dispatch_model
+
+        model = AutoModel.from_pretrained(checkpoint_path, trust_remote_code=True, **kwargs).half()
+
+        if device_map is None:
+            device_map = auto_configure_device_map(num_gpus)
+
+        model = dispatch_model(model, device_map=device_map)
+
+    return model
+
+
+class InvalidScoreLogitsProcessor(LogitsProcessor):
+    def __call__(
+            self, input_ids: torch.LongTensor, scores: torch.FloatTensor
+    ) -> torch.FloatTensor:
+        if torch.isnan(scores).any() or torch.isinf(scores).any():
+            scores.zero_()
+            scores[..., 5] = 5e4
+        return scores
+
+
+def process_response(output: str, use_tool: bool = False) -> Union[str, dict]:
+    content = ""
+    for response in output.split("<|assistant|>"):
+        metadata, content = response.split("\n", maxsplit=1)
+        if not metadata.strip():
+            content = content.strip()
+            content = content.replace("[[训练时间]]", "2023年")
+        else:
+            if use_tool:
+                content = "\n".join(content.split("\n")[1:-1])
+
+                def tool_call(**kwargs):
+                    return kwargs
+
+                parameters = eval(content)
+                content = {
+                    "name": metadata.strip(),
+                    "arguments": json.dumps(parameters, ensure_ascii=False)
+                }
+            else:
+                content = {
+                    "name": metadata.strip(),
+                    "content": content
+                }
+    return content
+
+
+@torch.inference_mode()
+def generate_stream_chatglm3(model: PreTrainedModel, tokenizer: PreTrainedTokenizer, params: dict):
+    messages = params["messages"]
+    functions = params["functions"]
+    temperature = float(params.get("temperature", 1.0))
+    repetition_penalty = float(params.get("repetition_penalty", 1.0))
+    top_p = float(params.get("top_p", 1.0))
+    max_new_tokens = int(params.get("max_tokens", 256))
+    echo = params.get("echo", True)
+    messages = process_chatglm_messages(messages, functions=functions)
+    query, role = messages[-1]["content"], messages[-1]["role"]
+
+    inputs = tokenizer.build_chat_input(query, history=messages[:-1], role=role)
+    inputs = inputs.to(model.device)
+    input_echo_len = len(inputs["input_ids"][0])
+
+    if input_echo_len >= model.config.seq_length:
+        print(f"Input length larger than {model.config.seq_length}")
+
+    eos_token_id = [
+        tokenizer.eos_token_id,
+        tokenizer.get_command("<|user|>"),
+    ]
+
+    gen_kwargs = {
+        "max_new_tokens": max_new_tokens,
+        "do_sample": True if temperature > 1e-5 else False,
+        "top_p": top_p,
+        "repetition_penalty": repetition_penalty,
+        "logits_processor": [InvalidScoreLogitsProcessor()],
+    }
+    if temperature > 1e-5:
+        gen_kwargs["temperature"] = temperature
+
+    total_len = 0
+    for total_ids in model.stream_generate(**inputs, eos_token_id=eos_token_id, **gen_kwargs):
+        total_ids = total_ids.tolist()[0]
+        total_len = len(total_ids)
+        if echo:
+            output_ids = total_ids[:-1]
+        else:
+            output_ids = total_ids[input_echo_len:-1]
+
+        response = tokenizer.decode(output_ids)
+        if response and response[-1] != "�":
+            response, stop_found = apply_stopping_strings(response, ["<|observation|>"])
+
+            yield {
+                "text": response,
+                "usage": {
+                    "prompt_tokens": input_echo_len,
+                    "completion_tokens": total_len - input_echo_len,
+                    "total_tokens": total_len,
+                },
+                "finish_reason": "function_call" if stop_found else None,
+            }
+
+            if stop_found:
+                break
+
+    # Only last stream result contains finish_reason, we set finish_reason as stop
+    ret = {
+        "text": response,
+        "usage": {
+            "prompt_tokens": input_echo_len,
+            "completion_tokens": total_len - input_echo_len,
+            "total_tokens": total_len,
+        },
+        "finish_reason": "stop",
+    }
+    yield ret
+
+    gc.collect()
+    torch.cuda.empty_cache()
+
+
+def process_chatglm_messages(messages, functions=None):
+    _messages = messages
+    messages = []
+    if functions:
+        messages.append(
+            {
+                "role": "system",
+                "content": "Answer the following questions as best as you can. You have access to the following tools:",
+                "tools": functions
+            }
+        )
+
+    for m in _messages:
+        role, content, func_call = m.role, m.content, m.function_call
+        if role == "function":
+            messages.append(
+                {
+                    "role": "observation",
+                    "content": content
+                }
+            )
+
+        elif role == "assistant" and func_call is not None:
+            for response in content.split("<|assistant|>"):
+                metadata, sub_content = response.split("\n", maxsplit=1)
+                messages.append(
+                    {
+                        "role": role,
+                        "metadata": metadata,
+                        "content": sub_content.strip()
+                    }
+                )
+        else:
+            messages.append({"role": role, "content": content})
+    return messages
+
+
+def generate_chatglm3(model: PreTrainedModel, tokenizer: PreTrainedTokenizer, params: dict):
+    for response in generate_stream_chatglm3(model, tokenizer, params):
+        pass
+    return response
+
+
+def apply_stopping_strings(reply, stop_strings) -> Tuple[str, bool]:
+    stop_found = False
+    for string in stop_strings:
+        idx = reply.find(string)
+        if idx != -1:
+            reply = reply[:idx]
+            stop_found = True
+            break
+
+    if not stop_found:
+        # If something like "\nYo" is generated just before "\nYou: is completed, trim it
+        for string in stop_strings:
+            for j in range(len(string) - 1, 0, -1):
+                if reply[-j:] == string[:j]:
+                    reply = reply[:-j]
+                    break
+            else:
+                continue
+
+            break
+
+    return reply, stop_found
diff --git a/requirements.txt b/requirements.txt
new file mode 100644
index 0000000000000000000000000000000000000000..beab1fe67c9b0678439d3de91508d59eec4ffa5b
--- /dev/null
+++ b/requirements.txt
@@ -0,0 +1,13 @@
+protobuf
+transformers>=4.30.2
+cpm_kernels
+torch>=2.0
+gradio~=3.39
+sentencepiece
+accelerate
+sse-starlette
+streamlit>=1.24.0
+fastapi>=0.104.1
+uvicorn~=0.24.0
+loguru~=0.7.2
+latex2mathml
\ No newline at end of file
diff --git a/resources/WECHAT.md b/resources/WECHAT.md
new file mode 100644
index 0000000000000000000000000000000000000000..c9ee867ead5d818a0b4e2ba46103a6454537d143
--- /dev/null
+++ b/resources/WECHAT.md
@@ -0,0 +1,7 @@
+<div align="center">
+<img src=wechat.jpg width="60%"/>
+
+<p> 扫码关注公众号，加入「ChatGLM交流群」 </p>
+<p> Scan the QR code to follow the official account and join the "ChatGLM Discussion Group" </p>
+</div>
+
diff --git a/resources/cli-demo.png b/resources/cli-demo.png
new file mode 100644
index 0000000000000000000000000000000000000000..3d489b5bd6d7e9d6b1db198307b8204f6fc2f80e
Binary files /dev/null and b/resources/cli-demo.png differ
diff --git a/resources/code_en.gif b/resources/code_en.gif
new file mode 100644
index 0000000000000000000000000000000000000000..d67baf5e3427f62c1233b45189f767dfaff2ad61
--- /dev/null
+++ b/resources/code_en.gif
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb3e6c3d236b5313ea5269d75b5bd49e3c92584675450858b43bbddedc0727d3
+size 9340751
diff --git a/resources/heart.png b/resources/heart.png
new file mode 100644
index 0000000000000000000000000000000000000000..0ee1462c14d2aab783ad66a1eccf451308514f2e
Binary files /dev/null and b/resources/heart.png differ
diff --git a/resources/tool.png b/resources/tool.png
new file mode 100644
index 0000000000000000000000000000000000000000..29bca13e8c93d57631d2a2e6c2ea68dd2699d8a9
Binary files /dev/null and b/resources/tool.png differ
diff --git a/resources/tool_en.png b/resources/tool_en.png
new file mode 100644
index 0000000000000000000000000000000000000000..af8a353a711bf5852aab9662b74878348a584d98
Binary files /dev/null and b/resources/tool_en.png differ
diff --git a/resources/web-demo.gif b/resources/web-demo.gif
new file mode 100644
index 0000000000000000000000000000000000000000..c775716f5a9608281dc3306022249afa121d5105
--- /dev/null
+++ b/resources/web-demo.gif
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba8ff042bbd879cbb4dd3795081b2e4e3713d3a4d2d5d7d61a027c389324cbbc
+size 2284452
diff --git a/resources/web-demo2.gif b/resources/web-demo2.gif
new file mode 100644
index 0000000000000000000000000000000000000000..00276843766e6475a1381ab1ad723bd0e65d9ba1
--- /dev/null
+++ b/resources/web-demo2.gif
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:999903f3fab1f3a09077fc932d4ff8ca4a77113274c444325ae556157e099632
+size 2761597
diff --git a/resources/web-demo2.png b/resources/web-demo2.png
new file mode 100644
index 0000000000000000000000000000000000000000..ff6dce6ed6418123f32c56de09ac06e1ddf25b1d
Binary files /dev/null and b/resources/web-demo2.png differ
diff --git a/resources/wechat.jpg b/resources/wechat.jpg
new file mode 100644
index 0000000000000000000000000000000000000000..8ea08f2423d294a9c50efad144fdc1058b079fde
Binary files /dev/null and b/resources/wechat.jpg differ
diff --git a/tool_using/README.md b/tool_using/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..3a3eb9ea78e58b8479d45ffedb8092b7f29c25df
--- /dev/null
+++ b/tool_using/README.md
@@ -0,0 +1,74 @@
+# 工具调用
+本文档将介绍如何使用 ChatGLM3-6B 进行工具调用。目前只有 ChatGLM3-6B 模型支持工具调用，而 ChatGLM3-6B-Base 和 ChatGLM3-6B-32K 模型不支持。
+
+## 构建 System Prompt
+这里以两个工具调用为例，首先准备好要构建的数据的描述信息。
+
+```python
+tools = [
+    {
+        "name": "track",
+        "description": "追踪指定股票的实时价格",
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "symbol": {
+                    "description": "需要追踪的股票代码"
+                }
+            },
+            "required": ['symbol']
+        }
+    },
+    {
+        "name": "text-to-speech",
+        "description": "将文本转换为语音",
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "text": {
+                    "description": "需要转换成语音的文本"
+                },
+                "voice": {
+                    "description": "要使用的语音类型（男声、女声等）"
+                },
+                "speed": {
+                    "description": "语音的速度（快、中等、慢等）"
+                }
+            },
+            "required": ['text']
+        }
+    }
+]
+system_info = {"role": "system", "content": "Answer the following questions as best as you can. You have access to the following tools:", "tools": tools}
+```
+请确保工具的定义格式与例子中一致以获得最优的性能
+
+## 提出问题
+注意：目前 ChatGLM3-6B 的工具调用只支持通过 `chat` 方法，不支持  `stream_chat` 方法。
+```python
+history = [system_info]
+query = "帮我查询股票10111的价格"
+response, history = model.chat(tokenizer, query, history=history)
+print(response)
+```
+这里期望得到的输出为
+```json
+{"name": "track", "parameters": {"symbol": "10111"}}
+```
+这表示模型需要调用工具 `track`，并且需要传入参数 `symbol`。
+
+## 调用工具，生成回复
+这里需要自行实现调用工具的逻辑。假设已经得到了返回结果，将结果以 json 格式返回给模型并得到回复。
+```python
+result = json.dumps({"price": 12412}, ensure_ascii=False)
+response, history = model.chat(tokenizer, result, history=history, role="observation")
+print(response)
+```
+这里 `role="observation"` 表示输入的是工具调用的返回值而不是用户输入，不能省略。
+
+期望得到的输出为
+```
+根据您的查询，经过API的调用，股票10111的价格是12412。
+```
+
+这表示本次工具调用已经结束，模型根据返回结果生成回复。对于比较复杂的问题，模型可能需要进行多次工具调用。这时，可以根据返回的 `response` 是 `str` 还是 `dict` 来判断返回的是生成的回复还是工具调用请求。
\ No newline at end of file
diff --git a/tool_using/README_en.md b/tool_using/README_en.md
new file mode 100644
index 0000000000000000000000000000000000000000..0e53479e781ac878629526b156b1ee98404b318f
--- /dev/null
+++ b/tool_using/README_en.md
@@ -0,0 +1,75 @@
+# Tool Invocation
+This document will introduce how to use the ChatGLM3-6B for tool invocation. Currently, only the ChatGLM3-6B model supports tool invocation, while the ChatGLM3-6B-Base and ChatGLM3-6B-32K models do not support it.
+
+## Building System Prompt
+Here are two examples of tool invocation. First, prepare the description information of the data to be built.
+
+```python
+tools = [
+    {
+        "name": "track",
+        "description": "Track the real-time price of a specified stock",
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "symbol": {
+                    "description": "The stock code that needs to be tracked"
+                }
+            },
+            "required": ['symbol']
+        }
+    },
+    {
+        "name": "text-to-speech",
+        "description": "Convert text to speech",
+        "parameters": {
+            "type": "object",
+            "properties": {
+                "text": {
+                    "description": "The text that needs to be converted into speech"
+                },
+                "voice": {
+                    "description": "The type of voice to use (male, female, etc.)"
+                },
+                "speed": {
+                    "description": "The speed of the speech (fast, medium, slow, etc.)"
+                }
+            },
+            "required": ['text']
+        }
+    }
+]
+system_info = {"role": "system", "content": "Answer the following questions as best as you can. You have access to the following tools:", "tools": tools}
+```
+
+Please ensure that the definition format of the tool is consistent with the example to obtain optimal performance.
+
+## Asking Questions
+Note: Currently, the tool invocation of ChatGLM3-6B only supports the `chat` method and does not support the `stream_chat` method.
+```python
+history = [system_info]
+query = "Help me inquire the price of stock 10111"
+response, history = model.chat(tokenizer, query, history=history)
+print(response)
+```
+The expected output here is
+```json
+{"name": "track", "parameters": {"symbol": "10111"}}
+```
+This indicates that the model needs to call the tool `track`, and the parameter `symbol` needs to be passed in.
+
+## Invoke Tool, Generate Response
+Here, you need to implement the logic of calling the tool yourself. Assuming that the return result has been obtained, return the result to the model in json format and get a response.
+```python
+result = json.dumps({"price": 12412}, ensure_ascii=False)
+response, history = model.chat(tokenizer, result, history=history, role="observation")
+print(response)
+```
+Here `role="observation"` indicates that the input is the return value of the tool invocation rather than user input, and it cannot be omitted.
+
+The expected output is
+```
+Based on your query, after the API call, the price of stock 10111 is 12412.
+```
+
+This indicates that this tool invocation has ended, and the model generates a response based on the return result. For more complex questions, the model may need to make multiple tool invocations. At this time, you can judge whether the returned `response` is `str` or `dict` to determine whether the return is a generated response or a tool invocation request.
\ No newline at end of file
diff --git a/tool_using/cli_demo_tool.py b/tool_using/cli_demo_tool.py
new file mode 100644
index 0000000000000000000000000000000000000000..62b8bf729b06bc6e0e7bd8de7a6c367b7399274e
--- /dev/null
+++ b/tool_using/cli_demo_tool.py
@@ -0,0 +1,58 @@
+import os
+import platform
+from transformers import AutoTokenizer, AutoModel
+import torch
+
+MODEL_PATH = os.environ.get('MODEL_PATH', 'THUDM/chatglm3-6b')
+TOKENIZER_PATH = os.environ.get("TOKENIZER_PATH", MODEL_PATH)
+DEVICE = 'cuda' if torch.cuda.is_available() else 'cpu'
+
+tokenizer = AutoTokenizer.from_pretrained(TOKENIZER_PATH, trust_remote_code=True)
+if 'cuda' in DEVICE: # AMD, NVIDIA GPU can use Half Precision
+    model = AutoModel.from_pretrained(MODEL_PATH, trust_remote_code=True).to(DEVICE).eval()
+else: # CPU, Intel GPU and other GPU can use Float16 Precision Only
+    model = AutoModel.from_pretrained(MODEL_PATH, trust_remote_code=True).float().to(DEVICE).eval()
+
+os_name = platform.system()
+clear_command = 'cls' if os_name == 'Windows' else 'clear'
+stop_stream = False
+
+
+def build_prompt(history):
+    prompt = "欢迎使用 ChatGLM3-6B 模型，输入内容即可进行对话，clear 清空对话历史，stop 终止程序"
+    for query, response in history:
+        prompt += f"\n\n用户：{query}"
+        prompt += f"\n\nChatGLM3-6B：{response}"
+    return prompt
+
+
+tools = [{'name': 'track', 'description': '追踪指定股票的实时价格', 'parameters': {'type': 'object', 'properties': {'symbol': {'description': '需要追踪的股票代码'}}, 'required': []}}, {'name': '/text-to-speech', 'description': '将文本转换为语音', 'parameters': {'type': 'object', 'properties': {'text': {'description': '需要转换成语音的文本'}, 'voice': {'description': '要使用的语音类型（男声、女声等）'}, 'speed': {'description': '语音的速度（快、中等、慢等）'}}, 'required': []}}, {'name': '/image_resizer', 'description': '调整图片的大小和尺寸', 'parameters': {'type': 'object', 'properties': {'image_file': {'description': '需要调整大小的图片文件'}, 'width': {'description': '需要调整的宽度值'}, 'height': {'description': '需要调整的高度值'}}, 'required': []}}, {'name': '/foodimg', 'description': '通过给定的食品名称生成该食品的图片', 'parameters': {'type': 'object', 'properties': {'food_name': {'description': '需要生成图片的食品名称'}}, 'required': []}}]
+system_item = {"role": "system",
+               "content": "Answer the following questions as best as you can. You have access to the following tools:",
+               "tools": tools}
+
+def main():
+    past_key_values, history = None, [system_item]
+    role = "user"
+    global stop_stream
+    print("欢迎使用 ChatGLM3-6B 模型，输入内容即可进行对话，clear 清空对话历史，stop 终止程序")
+    while True:
+        query = input("\n用户：") if role == "user" else input("\n结果：")
+        if query.strip() == "stop":
+            break
+        if query.strip() == "clear":
+            past_key_values, history = None,  [system_item]
+            role = "user"
+            os.system(clear_command)
+            print("欢迎使用 ChatGLM3-6B 模型，输入内容即可进行对话，clear 清空对话历史，stop 终止程序")
+            continue
+        print("\nChatGLM：", end="")
+        response, history = model.chat(tokenizer, query, history=history, role=role)
+        print(response, end="", flush=True)
+        print("")
+        if isinstance(response, dict):
+            role = "observation"
+
+
+if __name__ == "__main__":
+    main()
\ No newline at end of file
diff --git a/tool_using/openai_api_demo.py b/tool_using/openai_api_demo.py
new file mode 100644
index 0000000000000000000000000000000000000000..06d008af1dc08d7872ae44a9cf2cdc794c31960f
--- /dev/null
+++ b/tool_using/openai_api_demo.py
@@ -0,0 +1,93 @@
+import json
+
+from openai import OpenAI
+from colorama import init, Fore
+from loguru import logger
+
+from tool_register import get_tools, dispatch_tool
+
+init(autoreset=True)
+client = OpenAI(
+  base_url="http://127.0.0.1:8000/v1",
+  api_key = "xxx"
+)
+
+functions = get_tools()
+
+
+def run_conversation(query: str, stream=False, functions=None, max_retry=5):
+    params = dict(model="chatglm3", messages=[{"role": "user", "content": query}], stream=stream)
+    if functions:
+        params["functions"] = functions
+    response = client.chat.completions.create(**params)
+
+    for _ in range(max_retry):
+        if not stream:
+            if response.choices[0].message.function_call:
+                function_call = response.choices[0].message.function_call
+                logger.info(f"Function Call Response: {function_call.model_dump()}")
+
+                function_args = json.loads(function_call.arguments)
+                tool_response = dispatch_tool(function_call.name, function_args)
+                logger.info(f"Tool Call Response: {tool_response}")
+
+                params["messages"].append(response.choices[0].message)
+                params["messages"].append(
+                    {
+                        "role": "function",
+                        "name": function_call.name,
+                        "content": tool_response,  # 调用函数返回结果
+                    }
+                )
+            else:
+                reply = response.choices[0].message.content
+                logger.info(f"Final Reply: \n{reply}")
+                return
+
+        else:
+            output = ""
+            for chunk in response:
+                content = chunk.choices[0].delta.content or ""
+                print(Fore.BLUE + content, end="", flush=True)
+                output += content
+
+                if chunk.choices[0].finish_reason == "stop":
+                    return
+
+                elif chunk.choices[0].finish_reason == "function_call":
+                    print("\n")
+
+                    function_call = chunk.choices[0].delta.function_call
+                    logger.info(f"Function Call Response: {function_call.model_dump()}")
+
+                    function_args = json.loads(function_call.arguments)
+                    tool_response = dispatch_tool(function_call.name, function_args)
+                    logger.info(f"Tool Call Response: {tool_response}")
+
+                    params["messages"].append(
+                        {
+                            "role": "assistant",
+                            "content": output
+                        }
+                    )
+                    params["messages"].append(
+                        {
+                            "role": "function",
+                            "name": function_call.name,
+                            "content": tool_response,  # 调用函数返回结果
+                        }
+                    )
+
+                    break
+
+        response = client.chat.completions.create(**params)
+
+
+if __name__ == "__main__":
+    query = "你是谁"
+    run_conversation(query, stream=True)
+
+    logger.info("\n=========== next conversation ===========")
+
+    query = "帮我查询北京的天气怎么样"
+    run_conversation(query, functions=functions, stream=True)
diff --git a/tool_using/requirements.txt b/tool_using/requirements.txt
new file mode 100644
index 0000000000000000000000000000000000000000..03987af4c5e793b8189e078a7f2f2f1c7411546d
--- /dev/null
+++ b/tool_using/requirements.txt
@@ -0,0 +1 @@
+openai>=1.2.0
diff --git a/tool_using/tool_register.py b/tool_using/tool_register.py
new file mode 100644
index 0000000000000000000000000000000000000000..39ae2fc2049faf175a9ad40a581f6fb441eb6022
--- /dev/null
+++ b/tool_using/tool_register.py
@@ -0,0 +1,115 @@
+import inspect
+import traceback
+from copy import deepcopy
+from pprint import pformat
+from types import GenericAlias
+from typing import get_origin, Annotated
+
+_TOOL_HOOKS = {}
+_TOOL_DESCRIPTIONS = {}
+
+
+def register_tool(func: callable):
+    tool_name = func.__name__
+    tool_description = inspect.getdoc(func).strip()
+    python_params = inspect.signature(func).parameters
+    tool_params = []
+    for name, param in python_params.items():
+        annotation = param.annotation
+        if annotation is inspect.Parameter.empty:
+            raise TypeError(f"Parameter `{name}` missing type annotation")
+        if get_origin(annotation) != Annotated:
+            raise TypeError(f"Annotation type for `{name}` must be typing.Annotated")
+
+        typ, (description, required) = annotation.__origin__, annotation.__metadata__
+        typ: str = str(typ) if isinstance(typ, GenericAlias) else typ.__name__
+        if not isinstance(description, str):
+            raise TypeError(f"Description for `{name}` must be a string")
+        if not isinstance(required, bool):
+            raise TypeError(f"Required for `{name}` must be a bool")
+
+        tool_params.append({
+            "name": name,
+            "description": description,
+            "type": typ,
+            "required": required
+        })
+    tool_def = {
+        "name": tool_name,
+        "description": tool_description,
+        "params": tool_params
+    }
+
+    print("[registered tool] " + pformat(tool_def))
+    _TOOL_HOOKS[tool_name] = func
+    _TOOL_DESCRIPTIONS[tool_name] = tool_def
+
+    return func
+
+
+def dispatch_tool(tool_name: str, tool_params: dict) -> str:
+    if tool_name not in _TOOL_HOOKS:
+        return f"Tool `{tool_name}` not found. Please use a provided tool."
+    tool_call = _TOOL_HOOKS[tool_name]
+    try:
+        ret = tool_call(**tool_params)
+    except:
+        ret = traceback.format_exc()
+    return str(ret)
+
+
+def get_tools() -> dict:
+    return deepcopy(_TOOL_DESCRIPTIONS)
+
+
+# Tool Definitions
+
+@register_tool
+def random_number_generator(
+        seed: Annotated[int, 'The random seed used by the generator', True],
+        range: Annotated[tuple[int, int], 'The range of the generated numbers', True],
+) -> int:
+    """
+    Generates a random number x, s.t. range[0] <= x < range[1]
+    """
+    if not isinstance(seed, int):
+        raise TypeError("Seed must be an integer")
+    if not isinstance(range, tuple):
+        raise TypeError("Range must be a tuple")
+    if not isinstance(range[0], int) or not isinstance(range[1], int):
+        raise TypeError("Range must be a tuple of integers")
+
+    import random
+    return random.Random(seed).randint(*range)
+
+
+@register_tool
+def get_weather(
+        city_name: Annotated[str, 'The name of the city to be queried', True],
+) -> str:
+    """
+    Get the current weather for `city_name`
+    """
+
+    if not isinstance(city_name, str):
+        raise TypeError("City name must be a string")
+
+    key_selection = {
+        "current_condition": ["temp_C", "FeelsLikeC", "humidity", "weatherDesc", "observation_time"],
+    }
+    import requests
+    try:
+        resp = requests.get(f"https://wttr.in/{city_name}?format=j1")
+        resp.raise_for_status()
+        resp = resp.json()
+        ret = {k: {_v: resp[k][0][_v] for _v in v} for k, v in key_selection.items()}
+    except:
+        import traceback
+        ret = "Error encountered while fetching weather data!\n" + traceback.format_exc()
+
+    return str(ret)
+
+
+if __name__ == "__main__":
+    print(dispatch_tool("get_weather", {"city_name": "beijing"}))
+    print(get_tools())