')
+ gr.Markdown("# "+i18n("小原同学"))
+ gr.HTML(get_html("footer.html").format(
+ versions=versions_html()), elem_id="footer")
+ gr.Markdown(CHUANHU_DESCRIPTION, elem_id="description")
+
+ with gr.Box(elem_id="chuanhu-training"):
+ with gr.Row():
+ gr.Markdown("## "+i18n("训练"))
+ gr.HTML(get_html("close_btn.html").format(
+ obj="box"), elem_classes="close-btn")
+ with gr.Tabs(elem_id="chuanhu-training-tabs"):
+ with gr.Tab(label="OpenAI "+i18n("微调")):
+ openai_train_status = gr.Markdown(label=i18n("训练状态"), value=i18n(
+ "查看[使用介绍](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#微调-gpt-35)"))
+
+ with gr.Tab(label=i18n("准备数据集")):
+ dataset_preview_json = gr.JSON(
+ label=i18n("数据集预览"))
+ dataset_selection = gr.Files(label=i18n("选择数据集"), file_types=[
+ ".xlsx", ".jsonl"], file_count="single")
+ upload_to_openai_btn = gr.Button(
+ i18n("上传到OpenAI"), variant="primary", interactive=False)
+
+ with gr.Tab(label=i18n("训练")):
+ openai_ft_file_id = gr.Textbox(label=i18n(
+ "文件ID"), value="", lines=1, placeholder=i18n("上传到 OpenAI 后自动填充"))
+ openai_ft_suffix = gr.Textbox(label=i18n(
+ "模型名称后缀"), value="", lines=1, placeholder=i18n("可选，用于区分不同的模型"))
+ openai_train_epoch_slider = gr.Slider(label=i18n(
+ "训练轮数（Epochs）"), minimum=1, maximum=100, value=3, step=1, interactive=True)
+ openai_start_train_btn = gr.Button(
+ i18n("开始训练"), variant="primary", interactive=False)
+
+ with gr.Tab(label=i18n("状态")):
+ openai_status_refresh_btn = gr.Button(i18n("刷新状态"))
+ openai_cancel_all_jobs_btn = gr.Button(
+ i18n("取消所有任务"))
+ add_to_models_btn = gr.Button(
+ i18n("添加训练好的模型到模型列表"), interactive=False)
+
+ with gr.Box(elem_id="web-config", visible=False):
+ gr.HTML(get_html('web_config.html').format(
+ enableCheckUpdate_config=check_update,
+ hideHistoryWhenNotLoggedIn_config=hide_history_when_not_logged_in,
+ forView_i18n=i18n("仅供查看"),
+ deleteConfirm_i18n_pref=i18n("你真的要删除 "),
+ deleteConfirm_i18n_suff=i18n(" 吗？"),
+ usingLatest_i18n=i18n("您使用的就是最新版！"),
+ updatingMsg_i18n=i18n("正在尝试更新..."),
+ updateSuccess_i18n=i18n("更新成功，请重启本程序"),
+ updateFailure_i18n=i18n(
+ "更新失败，请尝试[手动更新](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#手动更新)"),
+ regenerate_i18n=i18n("重新生成"),
+ deleteRound_i18n=i18n("删除这轮问答"),
+ renameChat_i18n=i18n("重命名该对话"),
+ validFileName_i18n=i18n("请输入有效的文件名，不要包含以下特殊字符："),
+ ))
+ with gr.Box(elem_id="fake-gradio-components", visible=False):
+ updateChuanhuBtn = gr.Button(
+ visible=False, elem_classes="invisible-btn", elem_id="update-chuanhu-btn")
+ changeSingleSessionBtn = gr.Button(
+ visible=False, elem_classes="invisible-btn", elem_id="change-single-session-btn")
+ changeOnlineSearchBtn = gr.Button(
+ visible=False, elem_classes="invisible-btn", elem_id="change-online-search-btn")
+ historySelectBtn = gr.Button(
+ visible=False, elem_classes="invisible-btn", elem_id="history-select-btn") # Not used
+
+ # https://github.com/gradio-app/gradio/pull/3296
+
+ def create_greeting(request: gr.Request):
+ if hasattr(request, "username") and request.username: # is not None or is not ""
+ logging.info(f"Get User Name: {request.username}")
+ user_info, user_name = gr.Markdown.update(
+ value=f"User: {request.username}"), request.username
+ else:
+ user_info, user_name = gr.Markdown.update(
+ value=f"", visible=False), ""
+ current_model = get_model(
+ model_name=MODELS[DEFAULT_MODEL], access_key=my_api_key, user_name=user_name)[0]
+ if not hide_history_when_not_logged_in or user_name:
+ loaded_stuff = current_model.auto_load()
+ else:
+ loaded_stuff = [gr.update(), gr.update(), gr.Chatbot.update(label=MODELS[DEFAULT_MODEL]), current_model.single_turn, current_model.temperature, current_model.top_p, current_model.n_choices, current_model.stop_sequence, current_model.token_upper_limit, current_model.max_generation_token, current_model.presence_penalty, current_model.frequency_penalty, current_model.logit_bias, current_model.user_identifier]
+ return user_info, user_name, current_model, toggle_like_btn_visibility(DEFAULT_MODEL), *loaded_stuff, init_history_list(user_name)
+ demo.load(create_greeting, inputs=None, outputs=[
+ user_info, user_name, current_model, like_dislike_area, saveFileName, systemPromptTxt, chatbot, single_turn_checkbox, temperature_slider, top_p_slider, n_choices_slider, stop_sequence_txt, max_context_length_slider, max_generation_slider, presence_penalty_slider, frequency_penalty_slider, logit_bias_txt, user_identifier_txt, historySelectList], api_name="load")
+ chatgpt_predict_args = dict(
+ fn=predict,
+ inputs=[
+ current_model,
+ user_question,
+ chatbot,
+ use_streaming_checkbox,
+ use_websearch_checkbox,
+ index_files,
+ language_select_dropdown,
+ ],
+ outputs=[chatbot, status_display],
+ show_progress=True,
+ )
+
+ start_outputing_args = dict(
+ fn=start_outputing,
+ inputs=[],
+ outputs=[submitBtn, cancelBtn],
+ show_progress=True,
+ )
+
+ end_outputing_args = dict(
+ fn=end_outputing, inputs=[], outputs=[submitBtn, cancelBtn]
+ )
+
+ reset_textbox_args = dict(
+ fn=reset_textbox, inputs=[], outputs=[user_input]
+ )
+
+ transfer_input_args = dict(
+ fn=transfer_input, inputs=[user_input], outputs=[
+ user_question, user_input, submitBtn, cancelBtn], show_progress=True
+ )
+
+ get_usage_args = dict(
+ fn=billing_info, inputs=[current_model], outputs=[
+ usageTxt], show_progress=False
+ )
+
+ load_history_from_file_args = dict(
+ fn=load_chat_history,
+ inputs=[current_model, historySelectList],
+ outputs=[saveFileName, systemPromptTxt, chatbot, single_turn_checkbox, temperature_slider, top_p_slider, n_choices_slider, stop_sequence_txt, max_context_length_slider, max_generation_slider, presence_penalty_slider, frequency_penalty_slider, logit_bias_txt, user_identifier_txt],
+ )
+
+ refresh_history_args = dict(
+ fn=get_history_list, inputs=[user_name], outputs=[historySelectList]
+ )
+
+ auto_name_chat_history_args = dict(
+ fn=auto_name_chat_history,
+ inputs=[current_model, name_chat_method, user_question, chatbot, single_turn_checkbox],
+ outputs=[historySelectList],
+ show_progress=False,
+ )
+
+ # Chatbot
+ cancelBtn.click(interrupt, [current_model], [])
+
+ user_input.submit(**transfer_input_args).then(**
+ chatgpt_predict_args).then(**end_outputing_args).then(**auto_name_chat_history_args)
+ user_input.submit(**get_usage_args)
+
+ # user_input.submit(auto_name_chat_history, [current_model, user_question, chatbot, user_name], [historySelectList], show_progress=False)
+
+ submitBtn.click(**transfer_input_args).then(**chatgpt_predict_args,
+ api_name="predict").then(**end_outputing_args).then(**auto_name_chat_history_args)
+ submitBtn.click(**get_usage_args)
+
+ # submitBtn.click(auto_name_chat_history, [current_model, user_question, chatbot, user_name], [historySelectList], show_progress=False)
+
+ index_files.upload(handle_file_upload, [current_model, index_files, chatbot, language_select_dropdown], [
+ index_files, chatbot, status_display])
+ summarize_btn.click(handle_summarize_index, [
+ current_model, index_files, chatbot, language_select_dropdown], [chatbot, status_display])
+
+ emptyBtn.click(
+ reset,
+ inputs=[current_model, retain_system_prompt_checkbox],
+ outputs=[chatbot, status_display, historySelectList, systemPromptTxt, single_turn_checkbox, temperature_slider, top_p_slider, n_choices_slider, stop_sequence_txt, max_context_length_slider, max_generation_slider, presence_penalty_slider, frequency_penalty_slider, logit_bias_txt, user_identifier_txt],
+ show_progress=True,
+ _js='(a,b)=>{return clearChatbot(a,b);}',
+ )
+
+ retryBtn.click(**start_outputing_args).then(
+ retry,
+ [
+ current_model,
+ chatbot,
+ use_streaming_checkbox,
+ use_websearch_checkbox,
+ index_files,
+ language_select_dropdown,
+ ],
+ [chatbot, status_display],
+ show_progress=True,
+ ).then(**end_outputing_args)
+ retryBtn.click(**get_usage_args)
+
+ delFirstBtn.click(
+ delete_first_conversation,
+ [current_model],
+ [status_display],
+ )
+
+ delLastBtn.click(
+ delete_last_conversation,
+ [current_model, chatbot],
+ [chatbot, status_display],
+ show_progress=False
+ )
+
+ likeBtn.click(
+ like,
+ [current_model],
+ [status_display],
+ show_progress=False
+ )
+
+ dislikeBtn.click(
+ dislike,
+ [current_model],
+ [status_display],
+ show_progress=False
+ )
+
+ two_column.change(update_doc_config, [two_column], None)
+
+ # LLM Models
+ keyTxt.change(set_key, [current_model, keyTxt], [
+ user_api_key, status_display], api_name="set_key").then(**get_usage_args)
+ keyTxt.submit(**get_usage_args)
+ single_turn_checkbox.change(
+ set_single_turn, [current_model, single_turn_checkbox], None, show_progress=False)
+ model_select_dropdown.change(get_model, [model_select_dropdown, lora_select_dropdown, user_api_key, temperature_slider, top_p_slider, systemPromptTxt, user_name, current_model], [
+ current_model, status_display, chatbot, lora_select_dropdown, user_api_key, keyTxt], show_progress=True, api_name="get_model")
+ model_select_dropdown.change(toggle_like_btn_visibility, [model_select_dropdown], [
+ like_dislike_area], show_progress=False)
+ # model_select_dropdown.change(
+ # toggle_file_type, [model_select_dropdown], [index_files], show_progress=False)
+ lora_select_dropdown.change(get_model, [model_select_dropdown, lora_select_dropdown, user_api_key, temperature_slider,
+ top_p_slider, systemPromptTxt, user_name, current_model], [current_model, status_display, chatbot], show_progress=True)
+
+ # Template
+ systemPromptTxt.input(set_system_prompt, [
+ current_model, systemPromptTxt], None)
+ templateRefreshBtn.click(get_template_dropdown, None, [
+ templateFileSelectDropdown])
+ templateFileSelectDropdown.input(
+ load_template,
+ [templateFileSelectDropdown],
+ [promptTemplates, templateSelectDropdown],
+ show_progress=True,
+ )
+ templateSelectDropdown.change(
+ get_template_content,
+ [promptTemplates, templateSelectDropdown, systemPromptTxt],
+ [systemPromptTxt],
+ show_progress=True,
+ )
+
+ # S&L
+ renameHistoryBtn.click(
+ rename_chat_history,
+ [current_model, saveFileName, chatbot, user_name],
+ [historySelectList],
+ show_progress=True,
+ _js='(a,b,c,d)=>{return saveChatHistory(a,b,c,d);}'
+ )
+ exportMarkdownBtn.click(
+ export_markdown,
+ [current_model, saveFileName, chatbot],
+ [],
+ show_progress=True,
+ )
+ historyRefreshBtn.click(**refresh_history_args)
+ historyDeleteBtn.click(delete_chat_history, [current_model, historySelectList], [status_display, historySelectList, chatbot], _js='(a,b,c)=>{return showConfirmationDialog(a, b, c);}').then(
+ reset,
+ inputs=[current_model, retain_system_prompt_checkbox],
+ outputs=[chatbot, status_display, historySelectList, systemPromptTxt],
+ show_progress=True,
+ _js='(a,b)=>{return clearChatbot(a,b);}',
+ )
+ historySelectList.input(**load_history_from_file_args)
+ uploadFileBtn.upload(upload_chat_history, [current_model, uploadFileBtn], [
+ saveFileName, systemPromptTxt, chatbot, single_turn_checkbox, temperature_slider, top_p_slider, n_choices_slider, stop_sequence_txt, max_context_length_slider, max_generation_slider, presence_penalty_slider, frequency_penalty_slider, logit_bias_txt, user_identifier_txt]).then(**refresh_history_args)
+ historyDownloadBtn.click(None, [
+ user_name, historySelectList], None, _js='(a,b)=>{return downloadHistory(a,b,".json");}')
+ historyMarkdownDownloadBtn.click(None, [
+ user_name, historySelectList], None, _js='(a,b)=>{return downloadHistory(a,b,".md");}')
+ historySearchTextbox.input(
+ filter_history,
+ [user_name, historySearchTextbox],
+ [historySelectList]
+ )
+
+ # Train
+ dataset_selection.upload(handle_dataset_selection, dataset_selection, [
+ dataset_preview_json, upload_to_openai_btn, openai_train_status])
+ dataset_selection.clear(handle_dataset_clear, [], [
+ dataset_preview_json, upload_to_openai_btn])
+ upload_to_openai_btn.click(upload_to_openai, [dataset_selection], [
+ openai_ft_file_id, openai_train_status], show_progress=True)
+
+ openai_ft_file_id.change(lambda x: gr.update(interactive=True) if len(
+ x) > 0 else gr.update(interactive=False), [openai_ft_file_id], [openai_start_train_btn])
+ openai_start_train_btn.click(start_training, [
+ openai_ft_file_id, openai_ft_suffix, openai_train_epoch_slider], [openai_train_status])
+
+ openai_status_refresh_btn.click(get_training_status, [], [
+ openai_train_status, add_to_models_btn])
+ add_to_models_btn.click(add_to_models, [], [
+ model_select_dropdown, openai_train_status], show_progress=True)
+ openai_cancel_all_jobs_btn.click(
+ cancel_all_jobs, [], [openai_train_status], show_progress=True)
+
+ # Advanced
+ temperature_slider.input(
+ set_temperature, [current_model, temperature_slider], None, show_progress=False)
+ top_p_slider.input(set_top_p, [current_model, top_p_slider], None, show_progress=False)
+ n_choices_slider.input(
+ set_n_choices, [current_model, n_choices_slider], None, show_progress=False)
+ stop_sequence_txt.input(
+ set_stop_sequence, [current_model, stop_sequence_txt], None, show_progress=False)
+ max_context_length_slider.input(
+ set_token_upper_limit, [current_model, max_context_length_slider], None, show_progress=False)
+ max_generation_slider.input(
+ set_max_tokens, [current_model, max_generation_slider], None, show_progress=False)
+ presence_penalty_slider.input(
+ set_presence_penalty, [current_model, presence_penalty_slider], None, show_progress=False)
+ frequency_penalty_slider.input(
+ set_frequency_penalty, [current_model, frequency_penalty_slider], None, show_progress=False)
+ logit_bias_txt.input(
+ set_logit_bias, [current_model, logit_bias_txt], None, show_progress=False)
+ user_identifier_txt.input(set_user_identifier, [
+ current_model, user_identifier_txt], None, show_progress=False)
+
+ default_btn.click(
+ reset_default, [], [apihostTxt, proxyTxt, status_display], show_progress=True
+ )
+ # changeAPIURLBtn.click(
+ # change_api_host,
+ # [apihostTxt],
+ # [status_display],
+ # show_progress=True,
+ # )
+ # changeProxyBtn.click(
+ # change_proxy,
+ # [proxyTxt],
+ # [status_display],
+ # show_progress=True,
+ # )
+ # checkUpdateBtn.click(fn=None, _js='manualCheckUpdate')
+
+ # Invisible elements
+ updateChuanhuBtn.click(
+ update_chuanhu,
+ [],
+ [status_display],
+ show_progress=True,
+ )
+ changeSingleSessionBtn.click(
+ fn=lambda value: gr.Checkbox.update(value=value),
+ inputs=[single_turn_checkbox],
+ outputs=[single_turn_checkbox],
+ _js='(a)=>{return bgChangeSingleSession(a);}'
+ )
+ changeOnlineSearchBtn.click(
+ fn=lambda value: gr.Checkbox.update(value=value),
+ inputs=[use_websearch_checkbox],
+ outputs=[use_websearch_checkbox],
+ _js='(a)=>{return bgChangeOnlineSearch(a);}'
+ )
+ historySelectBtn.click( # This is an experimental feature... Not actually used.
+ fn=load_chat_history,
+ inputs=[current_model, historySelectList],
+ outputs=[saveFileName, systemPromptTxt, chatbot, single_turn_checkbox, temperature_slider, top_p_slider, n_choices_slider, stop_sequence_txt, max_context_length_slider, max_generation_slider, presence_penalty_slider, frequency_penalty_slider, logit_bias_txt, user_identifier_txt],
+ _js='(a,b)=>{return bgSelectHistory(a,b);}'
+ )
+
+
+logging.info(
+ colorama.Back.GREEN
+ + "\n小原同学的温馨提示：访问 http://localhost:7860 查看界面"
+ + colorama.Style.RESET_ALL
+)
+# 默认开启本地服务器，默认可以直接从IP访问，默认不创建公开分享链接
+demo.title = i18n("小原同学 🚀")
+
+if __name__ == "__main__":
+ reload_javascript()
+ demo.queue(concurrency_count=CONCURRENT_COUNT).launch(
+ allowed_paths=["history", "web_assets"],
+ server_name=server_name,
+ server_port=server_port,
+ share=share,
+ auth=auth_from_conf if authflag else None,
+ favicon_path="./web_assets/favicon.ico",
+ inbrowser=not dockerflag, # 禁止在docker下开启inbrowser
+ )
diff --git a/Dockerfile b/Dockerfile
new file mode 100644
index 0000000000000000000000000000000000000000..85d5045d5316ac160277af1e7d60afa823c0f953
--- /dev/null
+++ b/Dockerfile
@@ -0,0 +1,18 @@
+FROM python:3.9-slim-buster as builder
+RUN apt-get update \
+ && apt-get install -y build-essential \
+ && apt-get clean \
+ && rm -rf /var/lib/apt/lists/*
+COPY requirements.txt .
+COPY requirements_advanced.txt .
+RUN pip install --user --no-cache-dir -r requirements.txt
+# RUN pip install --user --no-cache-dir -r requirements_advanced.txt
+
+FROM python:3.9-slim-buster
+LABEL maintainer="iskoldt"
+COPY --from=builder /root/.local /root/.local
+ENV PATH=/root/.local/bin:$PATH
+COPY . /app
+WORKDIR /app
+ENV dockerrun=yes
+CMD ["python3", "-u", "ChuanhuChatbot.py","2>&1", "|", "tee", "/var/log/application.log"]
diff --git a/LICENSE b/LICENSE
new file mode 100644
index 0000000000000000000000000000000000000000..f288702d2fa16d3cdf0035b15a9fcbc552cd88e7
--- /dev/null
+++ b/LICENSE
@@ -0,0 +1,674 @@
+ GNU GENERAL PUBLIC LICENSE
+ Version 3, 29 June 2007
+
+ Copyright (C) 2007 Free Software Foundation, Inc. 川虎 Chat 🐯 Chuanhu Chat
+
+
+ 
+
+
+
+
+[](https://github.com/GaiZhenbiao/ChuanhuChatGPT/assets/51039745/0eee1598-c2fd-41c6-bda9-7b059a3ce6e7?autoplay=1)
+
+## 目录
+
+| [支持模型](#支持模型) | [使用技巧](#使用技巧) | [安装方式](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程) | [常见问题](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/常见问题) | [给作者买可乐🥤](#捐款) |
+| --- | --- | --- | --- | --- |
+
+## ✨ 5.0 重磅更新！
+
+
+
+
+New! 全新的用户界面！精致得不像 Gradio，甚至有毛玻璃效果！
+
+New! 适配了移动端（包括全面屏手机的挖孔/刘海），层级更加清晰。
+
+New! 历史记录移到左侧，使用更加方便。并且支持搜索（支持正则）、删除、重命名。
+
+New! 现在可以让大模型自动命名历史记录（需在设置或配置文件中开启）。
+
+New! 现在可以将 川虎Chat 作为 PWA 应用程序安装，体验更加原生！支持 Chrome/Edge/Safari 等浏览器。
+
+New! 图标适配各个平台，看起来更舒服。
+
+New! 支持 Finetune（微调） GPT 3.5！
+
+
+## 支持模型
+
+| API 调用模型 | 备注 | 本地部署模型 | 备注 |
+| :---: | --- | :---: | --- |
+| [ChatGPT(GPT-4)](https://chat.openai.com) | 支持微调 gpt-3.5 | [ChatGLM](https://github.com/THUDM/ChatGLM-6B) ([ChatGLM2](https://github.com/THUDM/ChatGLM2-6B)) |
+| [Azure OpenAI](https://azure.microsoft.com/en-us/products/ai-services/openai-service) | | [LLaMA](https://github.com/facebookresearch/llama) | 支持 Lora 模型
+| [Google PaLM](https://developers.generativeai.google/products/palm) | 不支持流式传输 | [StableLM](https://github.com/Stability-AI/StableLM)
+| [讯飞星火认知大模型](https://xinghuo.xfyun.cn) | | [MOSS](https://github.com/OpenLMLab/MOSS)
+| [Inspur Yuan 1.0](https://air.inspur.com/home) | | [通义千问](https://github.com/QwenLM/Qwen/tree/main)
+| [MiniMax](https://api.minimax.chat/) |
+| [XMChat](https://github.com/MILVLG/xmchat) | 不支持流式传输
+| [Midjourney](https://www.midjourney.com/) | 不支持流式传输
+| [Claude](https://www.anthropic.com/) |
+
+## 使用技巧
+
+### 💪 强力功能
+- **川虎助理**：类似 AutoGPT，全自动解决你的问题；
+- **在线搜索**：ChatGPT 的数据太旧？给 LLM 插上网络的翅膀；
+- **知识库**：让 ChatGPT 帮你量子速读！根据文件回答问题。
+- **本地部署LLM**：一键部署，获取属于你自己的大语言模型。
+
+### 🤖 System Prompt
+- 通过 System Prompt 设定前提条件，可以很有效地进行角色扮演；
+- 川虎Chat 预设了Prompt模板，点击`加载Prompt模板`，先选择 Prompt 模板集合，然后在下方选择想要的 Prompt。
+
+### 💬 基础对话
+- 如果回答不满意，可以使用 `重新生成` 按钮再试一次，或者直接 `删除这轮对话`;
+- 输入框支持换行，按 Shift + Enter即可；
+- 在输入框按 ↑ ↓ 方向键，可以在发送记录中快速切换；
+- 每次新建一个对话太麻烦，试试 `单论对话` 功能；
+- 回答气泡旁边的小按钮，不仅能 `一键复制`，还能 `查看Markdown原文`；
+- 指定回答语言，让 ChatGPT 固定以某种语言回答。
+
+### 📜 对话历史
+- 对话历史记录会被自动保存，不用担心问完之后找不到了；
+- 多用户历史记录隔离，除了你都看不到；
+- 重命名历史记录，方便日后查找；
+- New! 魔法般自动命名历史记录，让 LLM 理解对话内容，帮你自动为历史记录命名！
+- New! 搜索历史记录，支持正则表达式！
+
+### 🖼️ 小而美的体验
+- 自研 Small-and-Beautiful 主题，带给你小而美的体验；
+- 自动亮暗色切换，给你从早到晚的舒适体验；
+- 完美渲染 LaTeX / 表格 / 代码块，支持代码高亮；
+- New! 非线性动画、毛玻璃效果，精致得不像 Gradio！
+- New! 适配 Windows / macOS / Linux / iOS / Android，从图标到全面屏适配，给你最合适的体验！
+- New! 支持以 PWA应用程序 安装，体验更加原生！
+
+### 👨💻 极客功能
+- New! 支持 Fine-tune（微调）gpt-3.5！
+- 大量 LLM 参数可调；
+- 支持更换 api-host；
+- 支持自定义代理；
+- 支持多 api-key 负载均衡。
+
+### ⚒️ 部署相关
+- 部署到服务器：在 `config.json` 中设置 `"server_name": "0.0.0.0", "server_port": <你的端口号>,`。
+- 获取公共链接：在 `config.json` 中设置 `"share": true,`。注意程序必须在运行，才能通过公共链接访问。
+- 在Hugging Face上使用：建议在右上角 **复制Space** 再使用，这样App反应可能会快一点。
+
+## 快速上手
+
+在终端执行以下命令：
+
+```shell
+git clone https://github.com/GaiZhenbiao/ChuanhuChatGPT.git
+cd ChuanhuChatGPT
+pip install -r requirements.txt
+```
+
+然后，在项目文件夹中复制一份 `config_example.json`，并将其重命名为 `config.json`，在其中填入 `API-Key` 等设置。
+
+```shell
+python ChuanhuChatbot.py
+```
+
+一个浏览器窗口将会自动打开，此时您将可以使用 **川虎Chat** 与ChatGPT或其他模型进行对话。
+
+> **Note**
+>
+> 具体详尽的安装教程和使用教程请查看[本项目的wiki页面](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程)。
+
+## 疑难杂症解决
+
+在遇到各种问题查阅相关信息前，您可以先尝试 **手动拉取本项目的最新更改1** 并 **更新依赖库2**，然后重试。步骤为：
+
+1. 点击网页上的 `Download ZIP` 按钮，下载最新代码并解压覆盖，或
+ ```shell
+ git pull https://github.com/GaiZhenbiao/ChuanhuChatGPT.git main -f
+ ```
+2. 尝试再次安装依赖（可能本项目引入了新的依赖）
+ ```
+ pip install -r requirements.txt
+ ```
+
+很多时候，这样就可以解决问题。
+
+如果问题仍然存在，请查阅该页面：[常见问题](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/常见问题)
+
+该页面列出了**几乎所有**您可能遇到的各种问题，包括如何配置代理，以及遇到问题后您该采取的措施，**请务必认真阅读**。
+
+## 了解更多
+
+若需了解更多信息，请查看我们的 [wiki](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki)：
+
+- [想要做出贡献？](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/贡献指南)
+- [项目更新情况？](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/更新日志)
+- [二次开发许可？](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用许可)
+- [如何引用项目？](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用许可#如何引用该项目)
+
+## Starchart
+
+[](https://star-history.com/#GaiZhenbiao/ChuanhuChatGPT&Date)
+
+## Contributors
+
+
+ +
为ChatGPT等多种LLM提供了一个轻快好用的Web图形界面和众多附加功能
+
+ 
+ 
+ + 支持 GPT-4 · 基于文件问答 · LLM本地部署 · 联网搜索 · Agent 助理 · 支持 Fine-tune +
+ 视频教程 + · + 2.0介绍视频 + || + 在线体验 + · + 一键部署 + + +
+
+
diff --git a/config_example.json b/config_example.json
new file mode 100644
index 0000000000000000000000000000000000000000..64f8a57de79987f827eca7d8557db633d61328ad
--- /dev/null
+++ b/config_example.json
@@ -0,0 +1,85 @@
+{
+ // 各配置具体说明，见 [https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#配置-configjson]
+
+ //== API 配置 ==
+ "openai_api_key": "", // 你的 OpenAI API Key，一般必填，若空缺则需在图形界面中填入API Key
+ "google_palm_api_key": "", // 你的 Google PaLM API Key，用于 Google PaLM 对话模型
+ "xmchat_api_key": "", // 你的 xmchat API Key，用于 XMChat 对话模型
+ "minimax_api_key": "", // 你的 MiniMax API Key，用于 MiniMax 对话模型
+ "minimax_group_id": "", // 你的 MiniMax Group ID，用于 MiniMax 对话模型
+ "midjourney_proxy_api_base": "https://xxx/mj", // 你的 https://github.com/novicezk/midjourney-proxy 代理地址
+ "midjourney_proxy_api_secret": "", // 你的 MidJourney Proxy API Secret，用于鉴权访问 api，可选
+ "midjourney_discord_proxy_url": "", // 你的 MidJourney Discord Proxy URL，用于对生成对图进行反代，可选
+ "midjourney_temp_folder": "./tmp", // 你的 MidJourney 临时文件夹，用于存放生成的图片，填空则关闭自动下载切图（直接显示MJ的四宫格图）
+ "spark_appid": "", // 你的 讯飞星火大模型 API AppID，用于讯飞星火大模型对话模型
+ "spark_api_key": "", // 你的 讯飞星火大模型 API Key，用于讯飞星火大模型对话模型
+ "spark_api_secret": "", // 你的 讯飞星火大模型 API Secret，用于讯飞星火大模型对话模型
+ "claude_api_secret":"",// 你的 Claude API Secret，用于 Claude 对话模型
+ "ernie_api_key": "",// 你的文心一言在百度云中的API Key，用于文心一言对话模型
+ "ernie_secret_key": "",// 你的文心一言在百度云中的Secret Key，用于文心一言对话模型
+
+
+ //== Azure ==
+ "openai_api_type": "openai", // 可选项：azure, openai
+ "azure_openai_api_key": "", // 你的 Azure OpenAI API Key，用于 Azure OpenAI 对话模型
+ "azure_openai_api_base_url": "", // 你的 Azure Base URL
+ "azure_openai_api_version": "2023-05-15", // 你的 Azure OpenAI API 版本
+ "azure_deployment_name": "", // 你的 Azure OpenAI Chat 模型 Deployment 名称
+ "azure_embedding_deployment_name": "", // 你的 Azure OpenAI Embedding 模型 Deployment 名称
+ "azure_embedding_model_name": "text-embedding-ada-002", // 你的 Azure OpenAI Embedding 模型名称
+
+ //== 基础配置 ==
+ "language": "auto", // 界面语言，可选"auto", "zh_CN", "en_US", "ja_JP", "ko_KR", "sv_SE", "ru_RU", "vi_VN"
+ "users": [], // 用户列表，[[用户名1, 密码1], [用户名2, 密码2], ...]
+ "local_embedding": false, //是否在本地编制索引
+ "hide_history_when_not_logged_in": false, //未登录情况下是否不展示对话历史
+ "check_update": true, //是否启用检查更新
+ "default_model": "gpt-3.5-turbo", // 默认模型
+ "chat_name_method_index": 2, // 选择对话名称的方法。0: 使用日期时间命名；1: 使用第一条提问命名，2: 使用模型自动总结
+ "bot_avatar": "default", // 机器人头像，可填写本地或网络图片链接，或者"none"（不显示头像）
+ "user_avatar": "default", // 用户头像，可填写本地或网络图片链接，或者"none"（不显示头像）
+
+ //== API 用量 ==
+ "show_api_billing": false, //是否显示OpenAI API用量（启用需要填写sensitive_id）
+ "sensitive_id": "", // 你 OpenAI 账户的 Sensitive ID，用于查询 API 用量
+ "usage_limit": 120, // 该 OpenAI API Key 的当月限额，单位：美元，用于计算百分比和显示上限
+ "legacy_api_usage": false, // 是否使用旧版 API 用量查询接口（OpenAI现已关闭该接口，但是如果你在使用第三方 API，第三方可能仍然支持此接口）
+
+ //== 川虎助理设置 ==
+ "default_chuanhu_assistant_model": "gpt-4", //川虎助理使用的模型，可选gpt-3.5-turbo或者gpt-4等
+ "GOOGLE_CSE_ID": "", //谷歌搜索引擎ID，用于川虎助理Pro模式，获取方式请看 https://stackoverflow.com/questions/37083058/programmatically-searching-google-in-python-using-custom-search
+ "GOOGLE_API_KEY": "", //谷歌API Key，用于川虎助理Pro模式
+ "WOLFRAM_ALPHA_APPID": "", //Wolfram Alpha API Key，用于川虎助理Pro模式，获取方式请看 https://products.wolframalpha.com/api/
+ "SERPAPI_API_KEY": "", //SerpAPI API Key，用于川虎助理Pro模式，获取方式请看 https://serpapi.com/
+
+ //== 文档处理与显示 ==
+ "latex_option": "default", // LaTeX 公式渲染策略，可选"default", "strict", "all"或者"disabled"
+ "advance_docs": {
+ "pdf": {
+ "two_column": false, // 是否认为PDF是双栏的
+ "formula_ocr": true // 是否使用OCR识别PDF中的公式
+ }
+ },
+
+ //== 高级配置 ==
+ // 是否多个API Key轮换使用
+ "multi_api_key": false,
+ "hide_my_key": false, // 如果你想在UI中隐藏 API 密钥输入框，将此值设置为 true
+ // "available_models": ["GPT3.5 Turbo", "GPT4 Turbo", "GPT4 Vision"], // 可用的模型列表，将覆盖默认的可用模型列表
+ // "extra_models": ["模型名称3", "模型名称4", ...], // 额外的模型，将添加到可用的模型列表之后
+ // "api_key_list": [
+ // "sk-xxxxxxxxxxxxxxxxxxxxxxxx1",
+ // "sk-xxxxxxxxxxxxxxxxxxxxxxxx2",
+ // "sk-xxxxxxxxxxxxxxxxxxxxxxxx3"
+ // ],
+ // 自定义OpenAI API Base
+ // "openai_api_base": "https://api.openai.com",
+ // 自定义使用代理（请替换代理URL）
+ // "https_proxy": "http://127.0.0.1:1079",
+ // "http_proxy": "http://127.0.0.1:1079",
+ // 自定义端口、自定义ip（请替换对应内容）
+ // "server_name": "0.0.0.0",
+ // "server_port": 7860,
+ // 如果要share到gradio，设置为true
+ // "share": false,
+}
diff --git a/configs/ds_config_chatbot.json b/configs/ds_config_chatbot.json
new file mode 100644
index 0000000000000000000000000000000000000000..09b0b7ae082ff57d45b87bf6ee3662459b741def
--- /dev/null
+++ b/configs/ds_config_chatbot.json
@@ -0,0 +1,17 @@
+{
+ "fp16": {
+ "enabled": false
+ },
+ "bf16": {
+ "enabled": true
+ },
+ "comms_logger": {
+ "enabled": false,
+ "verbose": false,
+ "prof_all": false,
+ "debug": false
+ },
+ "steps_per_print": 20000000000000000,
+ "train_micro_batch_size_per_gpu": 1,
+ "wall_clock_breakdown": false
+}
diff --git a/locale/en_US.json b/locale/en_US.json
new file mode 100644
index 0000000000000000000000000000000000000000..849ee87a2ecfee3c7a44e49239ed4c2c9507000c
--- /dev/null
+++ b/locale/en_US.json
@@ -0,0 +1,144 @@
+{
+ " 吗？": " ?",
+ "# ⚠️ 务必谨慎更改 ⚠️": "# ⚠️ Caution: Changes require care. ⚠️",
+ "**发送消息** 或 **提交key** 以显示额度": "**Send message** or **Submit key** to display credit",
+ "**本月使用金额** ": "**Monthly usage** ",
+ "**获取API使用情况失败**": "**Failed to get API usage**",
+ "**获取API使用情况失败**，sensitive_id错误或已过期": "**Failed to get API usage**, wrong or expired sensitive_id",
+ "**获取API使用情况失败**，需在填写`config.json`中正确填写sensitive_id": "**Failed to get API usage**, correct sensitive_id needed in `config.json`",
+ "API key为空，请检查是否输入正确。": "API key is empty, check whether it is entered correctly.",
+ "API密钥更改为了": "The API key is changed to",
+ "JSON解析错误,收到的内容: ": "JSON parsing error, received content: ",
+ "SSL错误，无法获取对话。": "SSL error, unable to get dialogue.",
+ "Token 计数: ": "Token Count: ",
+ "☹️发生了错误：": "☹️Error: ",
+ "⚠️ 为保证API-Key安全，请在配置文件`config.json`中修改网络设置": "⚠️ To ensure the security of API-Key, please modify the network settings in the configuration file `config.json`.",
+ "。你仍然可以使用聊天功能。": ". You can still use the chat function.",
+ "上传": "Upload",
+ "上传了": "Uploaded",
+ "上传到 OpenAI 后自动填充": "Automatically filled after uploading to OpenAI",
+ "上传到OpenAI": "Upload to OpenAI",
+ "上传文件": "Upload files",
+ "仅供查看": "For viewing only",
+ "从Prompt模板中加载": "Load from Prompt Template",
+ "从列表中加载对话": "Load dialog from list",
+ "代理地址": "Proxy address",
+ "代理错误，无法获取对话。": "Proxy error, unable to get dialogue.",
+ "你没有权限访问 GPT4，[进一步了解](https://github.com/GaiZhenbiao/ChuanhuChatGPT/issues/843)": "You do not have permission to access GPT-4, [learn more](https://github.com/GaiZhenbiao/ChuanhuChatGPT/issues/843)",
+ "你没有选择任何对话历史": "You have not selected any conversation history.",
+ "你真的要删除 ": "Are you sure you want to delete ",
+ "使用在线搜索": "Use online search",
+ "停止符，用英文逗号隔开...": "Type in stop token here, separated by comma...",
+ "关于": "About",
+ "准备数据集": "Prepare Dataset",
+ "切换亮暗色主题": "Switch light/dark theme",
+ "删除对话历史成功": "Successfully deleted conversation history.",
+ "删除这轮问答": "Delete this round of Q&A",
+ "刷新状态": "Refresh Status",
+ "剩余配额不足，[进一步了解](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/%E5%B8%B8%E8%A7%81%E9%97%AE%E9%A2%98#you-exceeded-your-current-quota-please-check-your-plan-and-billing-details)": "Insufficient remaining quota, [learn more](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/%E5%B8%B8%E8%A7%81%E9%97%AE%E9%A2%98#you-exceeded-your-current-quota-please-check-your-plan-and-billing-details)",
+ "加载Prompt模板": "Load Prompt Template",
+ "单轮对话": "Single-turn",
+ "历史记录（JSON）": "History file (JSON)",
+ "参数": "Parameters",
+ "双栏pdf": "Two-column pdf",
+ "取消": "Cancel",
+ "取消所有任务": "Cancel All Tasks",
+ "可选，用于区分不同的模型": "Optional, used to distinguish different models",
+ "启用的工具：": "Enabled tools: ",
+ "在工具箱中管理知识库文件": "Manage knowledge base files in the toolbox",
+ "在线搜索": "Web search",
+ "在这里输入": "Type in here",
+ "在这里输入System Prompt...": "Type in System Prompt here...",
+ "多账号模式已开启，无需输入key，可直接开始对话": "Multi-account mode is enabled, no need to enter key, you can start the dialogue directly",
+ "好": "OK",
+ "实时传输回答": "Stream output",
+ "对话": "Dialogue",
+ "对话历史": "Conversation history",
+ "对话历史记录": "Dialog History",
+ "对话命名方式": "History naming method",
+ "导出为 Markdown": "Export as Markdown",
+ "川虎Chat": "川虎Chat",
+ "川虎Chat 🚀": "川虎Chat 🚀",
+ "小原同学": "Prinvest Mate",
+ "小原同学 🚀": "Prinvest Mate 🚀",
+ "工具箱": "Toolbox",
+ "已经被删除啦": "It has been deleted.",
+ "开始实时传输回答……": "Start streaming output...",
+ "开始训练": "Start Training",
+ "微调": "Fine-tuning",
+ "总结": "Summarize",
+ "总结完成": "Summary completed.",
+ "您使用的就是最新版！": "You are using the latest version!",
+ "您的IP区域：": "Your IP region: ",
+ "您的IP区域：未知。": "Your IP region: Unknown.",
+ "拓展": "Extensions",
+ "搜索（支持正则）...": "Search (supports regex)...",
+ "数据集预览": "Dataset Preview",
+ "文件ID": "File ID",
+ "新对话 ": "New Chat ",
+ "新建对话保留Prompt": "Retain Prompt For New Chat",
+ "暂时未知": "Unknown",
+ "更新": "Update",
+ "更新失败，请尝试[手动更新](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#手动更新)": "Update failed, please try [manually updating](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#手动更新)",
+ "更新成功，请重启本程序": "Updated successfully, please restart this program",
+ "未命名对话历史记录": "Unnamed Dialog History",
+ "未设置代理...": "No proxy...",
+ "本月使用金额": "Monthly usage",
+ "查看[使用介绍](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#微调-gpt-35)": "View the [usage guide](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#微调-gpt-35) for more details",
+ "根据日期时间": "By date and time",
+ "模型": "Model",
+ "模型名称后缀": "Model Name Suffix",
+ "模型自动总结（消耗tokens）": "Auto summary by LLM (Consume tokens)",
+ "模型设置为了：": "Model is set to: ",
+ "正在尝试更新...": "Trying to update...",
+ "添加训练好的模型到模型列表": "Add trained model to the model list",
+ "状态": "Status",
+ "生成内容总结中……": "Generating content summary...",
+ "用于定位滥用行为": "Used to locate abuse",
+ "用户标识符": "User identifier",
+ "由Bilibili [土川虎虎虎](https://space.bilibili.com/29125536)、[明昭MZhao](https://space.bilibili.com/24807452) 和 [Keldos](https://github.com/Keldos-Li) 开发访问川虎Chat的 [GitHub项目](https://github.com/GaiZhenbiao/ChuanhuChatGPT) 下载最新版脚本": "Developed by Bilibili [土川虎虎虎](https://space.bilibili.com/29125536), [明昭MZhao](https://space.bilibili.com/24807452) and [Keldos](https://github.com/Keldos-Li)\n\nDownload latest code from [GitHub](https://github.com/GaiZhenbiao/ChuanhuChatGPT)", + "知识库": "Knowledge base", + "知识库文件": "Knowledge base files", + "第一条提问": "By first question", + "索引构建完成": "Indexing complete.", + "网络": "Network", + "获取API使用情况失败:": "Failed to get API usage:", + "获取IP地理位置失败。原因：": "Failed to get IP location. Reason: ", + "获取对话时发生错误，请查看后台日志": "Error occurred when getting dialogue, check the background log", + "训练": "Training", + "训练状态": "Training Status", + "训练轮数（Epochs）": "Training Epochs", + "设置": "Settings", + "设置保存文件名": "Set save file name", + "设置文件名: 默认为.json，可选为.md": "Set file name: default is .json, optional is .md", + "识别公式": "formula OCR", + "详情": "Details", + "请查看 config_example.json，配置 Azure OpenAI": "Please review config_example.json to configure Azure OpenAI", + "请检查网络连接，或者API-Key是否有效。": "Check the network connection or whether the API-Key is valid.", + "请输入对话内容。": "Enter the content of the conversation.", + "请输入有效的文件名，不要包含以下特殊字符：": "Please enter a valid file name, do not include the following special characters: ", + "读取超时，无法获取对话。": "Read timed out, unable to get dialogue.", + "账单信息不适用": "Billing information is not applicable", + "连接超时，无法获取对话。": "Connection timed out, unable to get dialogue.", + "选择LoRA模型": "Select LoRA Model", + "选择Prompt模板集合文件": "Select Prompt Template Collection File", + "选择回复语言（针对搜索&索引功能）": "Select reply language (for search & index)", + "选择数据集": "Select Dataset", + "选择模型": "Select Model", + "重命名该对话": "Rename this chat", + "重新生成": "Regenerate", + "高级": "Advanced", + "，本次对话累计消耗了 ": ", total cost: ", + "💾 保存对话": "💾 Save Dialog", + "📝 导出为 Markdown": "📝 Export as Markdown", + "🔄 切换API地址": "🔄 Switch API Address", + "🔄 刷新": "🔄 Refresh", + "🔄 检查更新...": "🔄 Check for Update...", + "🔄 设置代理地址": "🔄 Set Proxy Address", + "🔄 重新生成": "🔄 Regeneration", + "🔙 恢复默认网络设置": "🔙 Reset Network Settings", + "🗑️ 删除最新对话": "🗑️ Delete latest dialog", + "🗑️ 删除最旧对话": "🗑️ Delete oldest dialog", + "🧹 新的对话": "🧹 New Dialogue", + "正在获取IP地址信息，请稍候...": "Getting IP address information, please wait...", +} \ No newline at end of file diff --git a/locale/extract_locale.py b/locale/extract_locale.py new file mode 100644 index 0000000000000000000000000000000000000000..dfbd90f6803099b9d1297fe807a4abf57f3abd91 --- /dev/null +++ b/locale/extract_locale.py @@ -0,0 +1,138 @@ +import os, json, re, sys +import aiohttp, asyncio +import commentjson + +asyncio.set_event_loop_policy(asyncio.DefaultEventLoopPolicy()) + +with open("config.json", "r", encoding="utf-8") as f: + config = commentjson.load(f) +api_key = config["openai_api_key"] +url = config["openai_api_base"] + "/chat/completions" if "openai_api_base" in config else "https://api.openai.com/v1/chat/completions" + + +def get_current_strings(): + pattern = r'i18n\s*\(\s*["\']([^"\']*(?:\)[^"\']*)?)["\']\s*\)' + + # Load the .py files + contents = "" + for dirpath, dirnames, filenames in os.walk("."): + for filename in filenames: + if filename.endswith(".py"): + filepath = os.path.join(dirpath, filename) + with open(filepath, 'r', encoding='utf-8') as f: + contents += f.read() + # Matching with regular expressions + matches = re.findall(pattern, contents, re.DOTALL) + data = {match.strip('()"'): '' for match in matches} + fixed_data = {} # fix some keys + for key, value in data.items(): + if "](" in key and key.count("(") != key.count(")"): + fixed_data[key+")"] = value + else: + fixed_data[key] = value + + return fixed_data + + +def get_locale_strings(filename): + try: + with open(filename, "r", encoding="utf-8") as f: + locale_strs = json.load(f) + except FileNotFoundError: + locale_strs = {} + return locale_strs + + +def sort_strings(existing_translations): + # Sort the merged data + sorted_translations = {} + # Add entries with (NOT USED) in their values + for key, value in sorted(existing_translations.items(), key=lambda x: x[0]): + if "(🔴NOT USED)" in value: + sorted_translations[key] = value + # Add entries with empty values + for key, value in sorted(existing_translations.items(), key=lambda x: x[0]): + if value == "": + sorted_translations[key] = value + # Add the rest of the entries + for key, value in sorted(existing_translations.items(), key=lambda x: x[0]): + if value != "" and "(NOT USED)" not in value: + sorted_translations[key] = value + + return sorted_translations + + +async def auto_translate(str, language): + headers = { + "Content-Type": "application/json", + "Authorization": f"Bearer {api_key}", + "temperature": f"{0}", + } + payload = { + "model": "gpt-3.5-turbo", + "messages": [ + { + "role": "system", + "content": f"You are a translation program;\nYour job is to translate user input into {language};\nThe content you are translating is a string in the App;\nDo not explain emoji;\nIf input is only a emoji, please simply return origin emoji;\nPlease ensure that the translation results are concise and easy to understand." + }, + {"role": "user", "content": f"{str}"} + ], + } + + async with aiohttp.ClientSession() as session: + async with session.post(url, headers=headers, json=payload) as response: + data = await response.json() + return data["choices"][0]["message"]["content"] + + +async def main(auto=False): + current_strs = get_current_strings() + locale_files = [] + # 遍历locale目录下的所有json文件 + for dirpath, dirnames, filenames in os.walk("locale"): + for filename in filenames: + if filename.endswith(".json"): + locale_files.append(os.path.join(dirpath, filename)) + + + for locale_filename in locale_files: + if "zh_CN" in locale_filename: + continue + locale_strs = get_locale_strings(locale_filename) + + # Add new keys + new_keys = [] + for key in current_strs: + if key not in locale_strs: + new_keys.append(key) + locale_strs[key] = "" + print(f"{locale_filename[7:-5]}'s new str: {len(new_keys)}") + # Add (NOT USED) to invalid keys + for key in locale_strs: + if key not in current_strs: + locale_strs[key] = "(🔴NOT USED)" + locale_strs[key] + print(f"{locale_filename[7:-5]}'s invalid str: {len(locale_strs) - len(current_strs)}") + + locale_strs = sort_strings(locale_strs) + + if auto: + tasks = [] + non_translated_keys = [] + for key in locale_strs: + if locale_strs[key] == "": + non_translated_keys.append(key) + tasks.append(auto_translate(key, locale_filename[7:-5])) + results = await asyncio.gather(*tasks) + for key, result in zip(non_translated_keys, results): + locale_strs[key] = "(🟡REVIEW NEEDED)" + result + print(f"{locale_filename[7:-5]}'s auto translated str: {len(non_translated_keys)}") + + with open(locale_filename, 'w', encoding='utf-8') as f: + json.dump(locale_strs, f, ensure_ascii=False, indent=4) + + +if __name__ == "__main__": + auto = False + if len(sys.argv) > 1 and sys.argv[1] == "--auto": + auto = True + asyncio.run(main(auto)) diff --git a/locale/ja_JP.json b/locale/ja_JP.json new file mode 100644 index 0000000000000000000000000000000000000000..70067a4ec06106fb8ee34223918fe504c2e58ede --- /dev/null +++ b/locale/ja_JP.json @@ -0,0 +1,144 @@ +{ + " 吗？": " を削除してもよろしいですか？", + "# ⚠️ 务必谨慎更改 ⚠️": "# ⚠️ 変更には慎重に ⚠️", + "**发送消息** 或 **提交key** 以显示额度": "**メッセージを送信** または **キーを送信** して、クレジットを表示します", + "**本月使用金额** ": "**今月の使用料金** ", + "**获取API使用情况失败**": "**API使用状況の取得に失敗しました**", + "**获取API使用情况失败**，sensitive_id错误或已过期": "**API使用状況の取得に失敗しました**、sensitive_idが間違っているか、期限切れです", + "**获取API使用情况失败**，需在填写`config.json`中正确填写sensitive_id": "**API使用状況の取得に失敗しました**、`config.json`に正しい`sensitive_id`を入力する必要があります", + "API key为空，请检查是否输入正确。": "APIキーが入力されていません。正しく入力されているか確認してください。", + "API密钥更改为了": "APIキーが変更されました", + "JSON解析错误,收到的内容: ": "JSON解析エラー、受信内容: ", + "SSL错误，无法获取对话。": "SSLエラー、会話を取得できません。", + "Token 计数: ": "Token数: ", + "☹️发生了错误：": "エラーが発生しました: ", + "⚠️ 为保证API-Key安全，请在配置文件`config.json`中修改网络设置": "⚠️ APIキーの安全性を確保するために、`config.json`ファイルでネットワーク設定を変更してください。", + "。你仍然可以使用聊天功能。": "。あなたはまだチャット機能を使用できます。", + "上传": "アップロード", + "上传了": "アップロードしました。", + "上传到 OpenAI 后自动填充": "OpenAIへのアップロード後、自動的に入力されます", + "上传到OpenAI": "OpenAIへのアップロード", + "上传文件": "ファイルをアップロード", + "仅供查看": "閲覧専用", + "从Prompt模板中加载": "Promptテンプレートから読込", + "从列表中加载对话": "リストから会話を読込", + "代理地址": "プロキシアドレス", + "代理错误，无法获取对话。": "プロキシエラー、会話を取得できません。", + "你没有权限访问 GPT4，[进一步了解](https://github.com/GaiZhenbiao/ChuanhuChatGPT/issues/843)": "GPT-4にアクセス権がありません、[詳細はこちら](https://github.com/GaiZhenbiao/ChuanhuChatGPT/issues/843)", + "你没有选择任何对话历史": "あなたは何の会話履歴も選択していません。", + "你真的要删除 ": "本当に ", + "使用在线搜索": "オンライン検索を使用", + "停止符，用英文逗号隔开...": "ここにストップ文字を英語のカンマで区切って入力してください...", + "关于": "について", + "准备数据集": "データセットの準備", + "切换亮暗色主题": "テーマの明暗切替", + "删除对话历史成功": "削除した会話の履歴", + "删除这轮问答": "この質疑応答を削除", + "刷新状态": "ステータスを更新", + "剩余配额不足，[进一步了解](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/%E5%B8%B8%E8%A7%81%E9%97%AE%E9%A2%98#you-exceeded-your-current-quota-please-check-your-plan-and-billing-details)": "剩余配额不足，[进一步了解]（https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/%E5%B8%B8%E8%A7%81%E9%97%AE%E9%A2%98#you-exceeded-your-current-quota-please-check-your-plan-and-billing-details）", + "加载Prompt模板": "Promptテンプレートを読込", + "单轮对话": "単発会話", + "历史记录（JSON）": "履歴ファイル（JSON）", + "参数": "パラメータ", + "双栏pdf": "2カラムpdf", + "取消": "キャンセル", + "取消所有任务": "すべてのタスクをキャンセル", + "可选，用于区分不同的模型": "オプション、異なるモデルを区別するために使用", + "启用的工具：": "有効なツール：", + "在工具箱中管理知识库文件": "ツールボックスでナレッジベースファイルの管理を行う", + "在线搜索": "オンライン検索", + "在这里输入": "ここに入力", + "在这里输入System Prompt...": "System Promptを入力してください...", + "多账号模式已开启，无需输入key，可直接开始对话": "複数アカウントモードがオンになっています。キーを入力する必要はありません。会話を開始できます", + "好": "はい", + "实时传输回答": "ストリーム出力", + "对话": "会話", + "对话历史": "対話履歴", + "对话历史记录": "会話履歴", + "对话命名方式": "会話の命名方法", + "导出为 Markdown": "Markdownでエクスポート", + "川虎Chat": "川虎Chat", + "川虎Chat 🚀": "川虎Chat 🚀", + "小原同学": "Prinvest Mate", + "小原同学 🚀": "Prinvest Mate 🚀", + "工具箱": "ツールボックス", + "已经被删除啦": "削除されました。", + "开始实时传输回答……": "ストリーム出力開始……", + "开始训练": "トレーニングを開始", + "微调": "ファインチューニング", + "总结": "要約する", + "总结完成": "完了", + "您使用的就是最新版！": "最新バージョンを使用しています！", + "您的IP区域：": "あなたのIPアドレス地域：", + "您的IP区域：未知。": "あなたのIPアドレス地域：不明", + "拓展": "拡張", + "搜索（支持正则）...": "検索（正規表現をサポート）...", + "数据集预览": "データセットのプレビュー", + "文件ID": "ファイルID", + "新对话 ": "新しい会話 ", + "新建对话保留Prompt": "新しい会話を作成してください。プロンプトを保留します。", + "暂时未知": "しばらく不明である", + "更新": "アップデート", + "更新失败，请尝试[手动更新](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#手动更新)": "更新に失敗しました、[手動での更新](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#手动更新)をお試しください。", + "更新成功，请重启本程序": "更新が成功しました、このプログラムを再起動してください", + "未命名对话历史记录": "名無しの会話履歴", + "未设置代理...": "代理が設定されていません...", + "本月使用金额": "今月の使用料金", + "查看[使用介绍](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#微调-gpt-35)": "[使用ガイド](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#微调-gpt-35)を表示", + "根据日期时间": "日付と時刻に基づいて", + "模型": "LLMモデル", + "模型名称后缀": "モデル名のサフィックス", + "模型自动总结（消耗tokens）": "モデルによる自動要約（トークン消費）", + "模型设置为了：": "LLMモデルを設定しました: ", + "正在尝试更新...": "更新を試みています...", + "添加训练好的模型到模型列表": "トレーニング済みモデルをモデルリストに追加", + "状态": "ステータス", + "生成内容总结中……": "コンテンツ概要を生成しています...", + "用于定位滥用行为": "不正行為を特定するために使用されます", + "用户标识符": "ユーザー識別子", + "由Bilibili [土川虎虎虎](https://space.bilibili.com/29125536)、[明昭MZhao](https://space.bilibili.com/24807452) 和 [Keldos](https://github.com/Keldos-Li) 开发
访问川虎Chat的 [GitHub项目](https://github.com/GaiZhenbiao/ChuanhuChatGPT) 下载最新版脚本": "開発：Bilibili [土川虎虎虎](https://space.bilibili.com/29125536) と [明昭MZhao](https://space.bilibili.com/24807452) と [Keldos](https://github.com/Keldos-Li)\n\n最新コードは川虎Chatのサイトへ [GitHubプロジェクト](https://github.com/GaiZhenbiao/ChuanhuChatGPT)", + "知识库": "ナレッジベース", + "知识库文件": "ナレッジベースファイル", + "第一条提问": "最初の質問", + "索引构建完成": "索引の構築が完了しました。", + "网络": "ネットワーク", + "获取API使用情况失败:": "API使用状況の取得に失敗しました:", + "获取IP地理位置失败。原因：": "IPアドレス地域の取得に失敗しました。理由：", + "获取对话时发生错误，请查看后台日志": "会話取得時にエラー発生、あとのログを確認してください", + "训练": "トレーニング", + "训练状态": "トレーニングステータス", + "训练轮数（Epochs）": "トレーニングエポック数", + "设置": "設定", + "设置保存文件名": "保存ファイル名を設定", + "设置文件名: 默认为.json，可选为.md": "ファイル名を設定: デフォルトは.json、.mdを選択できます", + "识别公式": "formula OCR", + "详情": "詳細", + "请查看 config_example.json，配置 Azure OpenAI": "Azure OpenAIの設定については、config_example.jsonをご覧ください", + "请检查网络连接，或者API-Key是否有效。": "ネットワーク接続を確認するか、APIキーが有効かどうかを確認してください。", + "请输入对话内容。": "会話内容を入力してください。", + "请输入有效的文件名，不要包含以下特殊字符：": "有効なファイル名を入力してください。以下の特殊文字は使用しないでください：", + "读取超时，无法获取对话。": "読み込みタイムアウト、会話を取得できません。", + "账单信息不适用": "課金情報は対象外です", + "连接超时，无法获取对话。": "接続タイムアウト、会話を取得できません。", + "选择LoRA模型": "LoRAモデルを選択", + "选择Prompt模板集合文件": "Promptテンプレートコレクションを選択", + "选择回复语言（针对搜索&索引功能）": "回答言語を選択（検索とインデックス機能に対して）", + "选择数据集": "データセットの選択", + "选择模型": "LLMモデルを選択", + "重命名该对话": "会話の名前を変更", + "重新生成": "再生成", + "高级": "Advanced", + "，本次对话累计消耗了 ": ", 今の会話で消費合計 ", + "💾 保存对话": "💾 会話を保存", + "📝 导出为 Markdown": "📝 Markdownにエクスポート", + "🔄 切换API地址": "🔄 APIアドレスを切り替え", + "🔄 刷新": "🔄 更新", + "🔄 检查更新...": "🔄 アップデートをチェック...", + "🔄 设置代理地址": "🔄 プロキシアドレスを設定", + "🔄 重新生成": "🔄 再生成", + "🔙 恢复默认网络设置": "🔙 ネットワーク設定のリセット", + "🗑️ 删除最新对话": "🗑️ 最新の会話削除", + "🗑️ 删除最旧对话": "🗑️ 最古の会話削除", + "🧹 新的对话": "🧹 新しい会話", + "正在获取IP地址信息，请稍候...": "IPアドレス情報を取得しています、しばらくお待ちください...", +} \ No newline at end of file diff --git a/locale/ko_KR.json b/locale/ko_KR.json new file mode 100644 index 0000000000000000000000000000000000000000..25637bdae4bac0209e37c76816f28b49962fa25e --- /dev/null +++ b/locale/ko_KR.json @@ -0,0 +1,144 @@ +{ + " 吗？": " 을(를) 삭제하시겠습니까?", + "# ⚠️ 务必谨慎更改 ⚠️": "# ⚠️ 주의: 변경시 주의하세요. ⚠️", + "**发送消息** 或 **提交key** 以显示额度": "**메세지를 전송** 하거나 **Key를 입력**하여 크레딧 표시", + "**本月使用金额** ": "**이번 달 사용금액** ", + "**获取API使用情况失败**": "**API 사용량 가져오기 실패**", + "**获取API使用情况失败**，sensitive_id错误或已过期": "**API 사용량 가져오기 실패**. sensitive_id가 잘못되었거나 만료되었습니다", + "**获取API使用情况失败**，需在填写`config.json`中正确填写sensitive_id": "**API 사용량 가져오기 실패**. `config.json`에 올바른 `sensitive_id`를 입력해야 합니다", + "API key为空，请检查是否输入正确。": "API 키가 비어 있습니다. 올바르게 입력되었는지 확인하십세요.", + "API密钥更改为了": "API 키가 변경되었습니다.", + "JSON解析错误,收到的内容: ": "JSON 파싱 에러, 응답: ", + "SSL错误，无法获取对话。": "SSL 에러, 대화를 가져올 수 없습니다.", + "Token 计数: ": "토큰 수: ", + "☹️发生了错误：": "☹️에러: ", + "⚠️ 为保证API-Key安全，请在配置文件`config.json`中修改网络设置": "⚠️ API-Key의 안전을 보장하기 위해 네트워크 설정을 `config.json` 구성 파일에서 수정해주세요.", + "。你仍然可以使用聊天功能。": ". 채팅 기능을 계속 사용할 수 있습니다.", + "上传": "업로드", + "上传了": "업로드완료.", + "上传到 OpenAI 后自动填充": "OpenAI로 업로드한 후 자동으로 채워집니다", + "上传到OpenAI": "OpenAI로 업로드", + "上传文件": "파일 업로드", + "仅供查看": "읽기 전용", + "从Prompt模板中加载": "프롬프트 템플릿에서 불러오기", + "从列表中加载对话": "리스트에서 대화 불러오기", + "代理地址": "프록시 주소", + "代理错误，无法获取对话。": "프록시 에러, 대화를 가져올 수 없습니다.", + "你没有权限访问 GPT4，[进一步了解](https://github.com/GaiZhenbiao/ChuanhuChatGPT/issues/843)": "GPT-4에 접근 권한이 없습니다. [자세히 알아보기](https://github.com/GaiZhenbiao/ChuanhuChatGPT/issues/843)", + "你没有选择任何对话历史": "대화 기록을 선택하지 않았습니다.", + "你真的要删除 ": "정말로 ", + "使用在线搜索": "온라인 검색 사용", + "停止符，用英文逗号隔开...": "여기에 정지 토큰 입력, ','로 구분됨...", + "关于": "관련", + "准备数据集": "데이터셋 준비", + "切换亮暗色主题": "라이트/다크 테마 전환", + "删除对话历史成功": "대화 기록이 성공적으로 삭제되었습니다.", + "删除这轮问答": "이 라운드의 질문과 답변 삭제", + "刷新状态": "상태 새로 고침", + "剩余配额不足，[进一步了解](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/%E5%B8%B8%E8%A7%81%E9%97%AE%E9%A2%98#you-exceeded-your-current-quota-please-check-your-plan-and-billing-details)": "남은 할당량이 부족합니다. [자세한 내용](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/%E5%B8%B8%E8%A7%81%E9%97%AE%E9%A2%98#you-exceeded-your-current-quota-please-check-your-plan-and-billing-details)을 확인하세요.", + "加载Prompt模板": "프롬프트 템플릿 불러오기", + "单轮对话": "단일 대화", + "历史记录（JSON）": "기록 파일 (JSON)", + "参数": "파라미터들", + "双栏pdf": "2-column pdf", + "取消": "취소", + "取消所有任务": "모든 작업 취소", + "可选，用于区分不同的模型": "선택 사항, 다른 모델을 구분하는 데 사용", + "启用的工具：": "활성화된 도구: ", + "在工具箱中管理知识库文件": "지식 라이브러리 파일을 도구 상자에서 관리", + "在线搜索": "온라인 검색", + "在这里输入": "여기에 입력하세요", + "在这里输入System Prompt...": "여기에 시스템 프롬프트를 입력하세요...", + "多账号模式已开启，无需输入key，可直接开始对话": "다중 계정 모드가 활성화되어 있으므로 키를 입력할 필요가 없이 바로 대화를 시작할 수 있습니다", + "好": "예", + "实时传输回答": "실시간 전송", + "对话": "대화", + "对话历史": "대화 내역", + "对话历史记录": "대화 기록", + "对话命名方式": "대화 이름 설정", + "导出为 Markdown": "Markdown으로 내보내기", + "川虎Chat": "Chuanhu Chat", + "川虎Chat 🚀": "Chuanhu Chat 🚀", + "小原同学": "Prinvest Mate", + "小原同学 🚀": "Prinvest Mate 🚀", + "工具箱": "도구 상자", + "已经被删除啦": "이미 삭제되었습니다.", + "开始实时传输回答……": "실시간 응답 출력 시작...", + "开始训练": "훈련 시작", + "微调": "파인튜닝", + "总结": "요약", + "总结完成": "작업 완료", + "您使用的就是最新版！": "최신 버전을 사용하고 있습니다!", + "您的IP区域：": "당신의 IP 지역: ", + "您的IP区域：未知。": "IP 지역: 알 수 없음.", + "拓展": "확장", + "搜索（支持正则）...": "검색 (정규식 지원)...", + "数据集预览": "데이터셋 미리보기", + "文件ID": "파일 ID", + "新对话 ": "새 대화 ", + "新建对话保留Prompt": "새 대화 생성, 프롬프트 유지하기", + "暂时未知": "알 수 없음", + "更新": "업데이트", + "更新失败，请尝试[手动更新](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#手动更新)": "업데이트 실패, [수동 업데이트](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#手动更新)를 시도하십시오", + "更新成功，请重启本程序": "업데이트 성공, 이 프로그램을 재시작 해주세요", + "未命名对话历史记录": "이름없는 대화 기록", + "未设置代理...": "프록시가 설정되지 않았습니다...", + "本月使用金额": "이번 달 사용금액", + "查看[使用介绍](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#微调-gpt-35)": "[사용 가이드](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#微调-gpt-35) 보기", + "根据日期时间": "날짜 및 시간 기준", + "模型": "LLM 모델", + "模型名称后缀": "모델 이름 접미사", + "模型自动总结（消耗tokens）": "모델에 의한 자동 요약 (토큰 소비)", + "模型设置为了：": "설정된 모델: ", + "正在尝试更新...": "업데이트를 시도 중...", + "添加训练好的模型到模型列表": "훈련된 모델을 모델 목록에 추가", + "状态": "상태", + "生成内容总结中……": "콘텐츠 요약 생성중...", + "用于定位滥用行为": "악용 사례 파악에 활용됨", + "用户标识符": "사용자 식별자", + "由Bilibili [土川虎虎虎](https://space.bilibili.com/29125536)、[明昭MZhao](https://space.bilibili.com/24807452) 和 [Keldos](https://github.com/Keldos-Li) 开发
访问川虎Chat的 [GitHub项目](https://github.com/GaiZhenbiao/ChuanhuChatGPT) 下载最新版脚本": "제작: Bilibili [土川虎虎虎](https://space.bilibili.com/29125536), [明昭MZhao](https://space.bilibili.com/24807452), [Keldos](https://github.com/Keldos-Li)\n\n최신 코드 다운로드: [GitHub](https://github.com/GaiZhenbiao/ChuanhuChatGPT)", + "知识库": "knowledge base", + "知识库文件": "knowledge base 파일", + "第一条提问": "첫 번째 질문", + "索引构建完成": "인덱스 구축이 완료되었습니다.", + "网络": "네트워크", + "获取API使用情况失败:": "API 사용량 가져오기 실패:", + "获取IP地理位置失败。原因：": "다음과 같은 이유로 IP 위치를 가져올 수 없습니다. 이유: ", + "获取对话时发生错误，请查看后台日志": "대화를 가져오는 중 에러가 발생했습니다. 백그라운드 로그를 확인하세요", + "训练": "학습", + "训练状态": "학습 상태", + "训练轮数（Epochs）": "학습 Epochs", + "设置": "설정", + "设置保存文件名": "저장 파일명 설정", + "设置文件名: 默认为.json，可选为.md": "파일 이름 설정: 기본값: .json, 선택: .md", + "识别公式": "formula OCR", + "详情": "상세", + "请查看 config_example.json，配置 Azure OpenAI": "Azure OpenAI 설정을 확인하세요", + "请检查网络连接，或者API-Key是否有效。": "네트워크 연결 또는 API키가 유효한지 확인하세요", + "请输入对话内容。": "대화 내용을 입력하세요.", + "请输入有效的文件名，不要包含以下特殊字符：": "유효한 파일 이름을 입력하세요. 다음 특수 문자를 포함하지 마세요: ", + "读取超时，无法获取对话。": "읽기 시간 초과, 대화를 가져올 수 없습니다.", + "账单信息不适用": "청구 정보를 가져올 수 없습니다", + "连接超时，无法获取对话。": "연결 시간 초과, 대화를 가져올 수 없습니다.", + "选择LoRA模型": "LoRA 모델 선택", + "选择Prompt模板集合文件": "프롬프트 콜렉션 파일 선택", + "选择回复语言（针对搜索&索引功能）": "답장 언어 선택 (검색 & 인덱스용)", + "选择数据集": "데이터셋 선택", + "选择模型": "모델 선택", + "重命名该对话": "대화 이름 변경", + "重新生成": "재생성", + "高级": "고급", + "，本次对话累计消耗了 ": "，이 대화의 전체 비용은 ", + "💾 保存对话": "💾 대화 저장", + "📝 导出为 Markdown": "📝 Markdown으로 내보내기", + "🔄 切换API地址": "🔄 API 주소 변경", + "🔄 刷新": "🔄 새로고침", + "🔄 检查更新...": "🔄 업데이트 확인...", + "🔄 设置代理地址": "🔄 프록시 주소 설정", + "🔄 重新生成": "🔄 재생성", + "🔙 恢复默认网络设置": "🔙 네트워크 설정 초기화", + "🗑️ 删除最新对话": "🗑️ 최신 대화 삭제", + "🗑️ 删除最旧对话": "🗑️ 가장 오래된 대화 삭제", + "🧹 新的对话": "🧹 새로운 대화", + "正在获取IP地址信息，请稍候...": "IP 주소 정보를 가져오는 중입니다. 잠시만 기다려주세요...", +} diff --git a/locale/ru_RU.json b/locale/ru_RU.json new file mode 100644 index 0000000000000000000000000000000000000000..4ffa73ec0a63a97d36cad43f30cd75d37d8b453f --- /dev/null +++ b/locale/ru_RU.json @@ -0,0 +1,144 @@ +{ + " 吗？": " ?", + "# ⚠️ 务必谨慎更改 ⚠️": "# ⚠️ ВНИМАНИЕ: ИЗМЕНЯЙТЕ ОСТОРОЖНО ⚠️", + "**发送消息** 或 **提交key** 以显示额度": "**Отправить сообщение** или **отправить ключ** для отображения лимита", + "**本月使用金额** ": "**Использовано средств в этом месяце**", + "**获取API使用情况失败**": "**Не удалось получить информацию об использовании API**", + "**获取API使用情况失败**，sensitive_id错误或已过期": "**Не удалось получить информацию об использовании API**, ошибка sensitive_id или истек срок действия", + "**获取API使用情况失败**，需在填写`config.json`中正确填写sensitive_id": "**Не удалось получить информацию об использовании API**, необходимо правильно заполнить sensitive_id в `config.json`", + "API key为空，请检查是否输入正确。": "Пустой API-Key, пожалуйста, проверьте правильность ввода.", + "API密钥更改为了": "Ключ API изменен на", + "JSON解析错误,收到的内容: ": "Ошибка анализа JSON, полученный контент:", + "SSL错误，无法获取对话。": "Ошибка SSL, не удалось получить диалог.", + "Token 计数: ": "Использованно токенов: ", + "☹️发生了错误：": "☹️ Произошла ошибка:", + "⚠️ 为保证API-Key安全，请在配置文件`config.json`中修改网络设置": "⚠️ Для обеспечения безопасности API-Key, измените настройки сети в файле конфигурации `config.json`", + "。你仍然可以使用聊天功能。": ". Вы все равно можете использовать функцию чата.", + "上传": "Загрузить", + "上传了": "Загрузка завершена.", + "上传到 OpenAI 后自动填充": "Автоматическое заполнение после загрузки в OpenAI", + "上传到OpenAI": "Загрузить в OpenAI", + "上传文件": "Загрузить файл", + "仅供查看": "Только для просмотра", + "从Prompt模板中加载": "Загрузить из шаблона Prompt", + "从列表中加载对话": "Загрузить диалог из списка", + "代理地址": "Адрес прокси", + "代理错误，无法获取对话。": "Ошибка прокси, не удалось получить диалог.", + "你没有权限访问 GPT4，[进一步了解](https://github.com/GaiZhenbiao/ChuanhuChatGPT/issues/843)": "У вас нет доступа к GPT4, [подробнее](https://github.com/GaiZhenbiao/ChuanhuChatGPT/issues/843)", + "你没有选择任何对话历史": "Вы не выбрали никакой истории переписки", + "你真的要删除 ": "Вы уверены, что хотите удалить ", + "使用在线搜索": "Использовать онлайн-поиск", + "停止符，用英文逗号隔开...": "Разделительные символы, разделенные запятой...", + "关于": "О программе", + "准备数据集": "Подготовка набора данных", + "切换亮暗色主题": "Переключить светлую/темную тему", + "删除对话历史成功": "Успешно удалена история переписки.", + "删除这轮问答": "Удалить этот раунд вопросов и ответов", + "刷新状态": "Обновить статус", + "剩余配额不足，[进一步了解](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/%E5%B8%B8%E8%A7%81%E9%97%AE%E9%A2%98#you-exceeded-your-current-quota-please-check-your-plan-and-billing-details)": "剩余配额不足，[进一步了解](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/%E5%B8%B8%E8%A7%81%E9%97%AE%E9%A2%98#you-exceeded-your-current-quota-please-check-your-plan-and-billing-details)", + "加载Prompt模板": "Загрузить шаблон Prompt", + "单轮对话": "Одиночный диалог", + "历史记录（JSON）": "Файл истории (JSON)", + "参数": "Параметры", + "双栏pdf": "Двухколоночный PDF", + "取消": "Отмена", + "取消所有任务": "Отменить все задачи", + "可选，用于区分不同的模型": "Необязательно, используется для различения разных моделей", + "启用的工具：": "Включенные инструменты：", + "在工具箱中管理知识库文件": "Управление файлами базы знаний в инструментах", + "在线搜索": "Онлайн-поиск", + "在这里输入": "Введите здесь", + "在这里输入System Prompt...": "Введите здесь системное подсказку...", + "多账号模式已开启，无需输入key，可直接开始对话": "Режим множественных аккаунтов включен, не требуется ввод ключа, можно сразу начать диалог", + "好": "Хорошо", + "实时传输回答": "Передача ответа в реальном времени", + "对话": "Диалог", + "对话历史": "Диалоговая история", + "对话历史记录": "История диалога", + "对话命名方式": "Способ названия диалога", + "导出为 Markdown": "Экспортировать в Markdown", + "川虎Chat": "Chuanhu Чат", + "川虎Chat 🚀": "Chuanhu Чат 🚀", + "小原同学": "Prinvest Mate", + "小原同学 🚀": "Prinvest Mate 🚀", + "工具箱": "Инструменты", + "已经被删除啦": "Уже удалено.", + "开始实时传输回答……": "Начните трансляцию ответов в режиме реального времени...", + "开始训练": "Начать обучение", + "微调": "Своя модель", + "总结": "Подведение итога", + "总结完成": "Готово", + "您使用的就是最新版！": "Вы используете последнюю версию!", + "您的IP区域：": "Ваша IP-зона:", + "您的IP区域：未知。": "Ваша IP-зона: неизвестно.", + "拓展": "Расширенные настройки", + "搜索（支持正则）...": "Поиск (поддержка регулярности)...", + "数据集预览": "Предпросмотр набора данных", + "文件ID": "Идентификатор файла", + "新对话 ": "Новый диалог ", + "新建对话保留Prompt": "Создать диалог с сохранением подсказки", + "暂时未知": "Временно неизвестно", + "更新": "Обновить", + "更新失败，请尝试[手动更新](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#手动更新)": "Обновление не удалось, пожалуйста, попробуйте обновить вручную", + "更新成功，请重启本程序": "Обновление успешно, пожалуйста, перезапустите программу", + "未命名对话历史记录": "Безымянная история диалога", + "未设置代理...": "Прокси не настроен...", + "本月使用金额": "Использовано средств в этом месяце", + "查看[使用介绍](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#微调-gpt-35)": "[Здесь](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#微调-gpt-35) можно ознакомиться с инструкцией по использованию", + "根据日期时间": "По дате и времени", + "模型": "Модель", + "模型名称后缀": "Суффикс имени модели", + "模型自动总结（消耗tokens）": "Автоматическое подведение итогов модели (потребление токенов)", + "模型设置为了：": "Модель настроена на:", + "正在尝试更新...": "Попытка обновления...", + "添加训练好的模型到模型列表": "Добавить обученную модель в список моделей", + "状态": "Статус", + "生成内容总结中……": "Создание сводки контента...", + "用于定位滥用行为": "Используется для выявления злоупотреблений", + "用户标识符": "Идентификатор пользователя", + "由Bilibili [土川虎虎虎](https://space.bilibili.com/29125536)、[明昭MZhao](https://space.bilibili.com/24807452) 和 [Keldos](https://github.com/Keldos-Li) 开发
访问川虎Chat的 [GitHub项目](https://github.com/GaiZhenbiao/ChuanhuChatGPT) 下载最新版脚本": "Разработано [土川虎虎虎](https://space.bilibili.com/29125536), [明昭MZhao](https://space.bilibili.com/24807452) и [Keldos](https://github.com/Keldos-Li).
посетите [GitHub Project](https://github.com/GaiZhenbiao/ChuanhuChatGPT) чата Chuanhu, чтобы загрузить последнюю версию скрипта", + "知识库": "База знаний", + "知识库文件": "Файл базы знаний", + "第一条提问": "Первый вопрос", + "索引构建完成": "Индексирование завершено.", + "网络": "Параметры сети", + "获取API使用情况失败:": "Не удалось получитьAPIинформацию об использовании:", + "获取IP地理位置失败。原因：": "Не удалось получить географическое положение IP. Причина:", + "获取对话时发生错误，请查看后台日志": "Возникла ошибка при получении диалога, пожалуйста, проверьте журналы", + "训练": "Обучение", + "训练状态": "Статус обучения", + "训练轮数（Epochs）": "Количество эпох обучения", + "设置": "Настройки", + "设置保存文件名": "Установить имя сохраняемого файла", + "设置文件名: 默认为.json，可选为.md": "Установить имя файла: по умолчанию .json, можно выбрать .md", + "识别公式": "Распознавание формул", + "详情": "Подробности", + "请查看 config_example.json，配置 Azure OpenAI": "Пожалуйста, просмотрите config_example.json для настройки Azure OpenAI", + "请检查网络连接，或者API-Key是否有效。": "Проверьте подключение к сети или действительность API-Key.", + "请输入对话内容。": "Пожалуйста, введите содержание диалога.", + "请输入有效的文件名，不要包含以下特殊字符：": "Введите действительное имя файла, не содержащее следующих специальных символов: ", + "读取超时，无法获取对话。": "Тайм-аут чтения, не удалось получить диалог.", + "账单信息不适用": "Информация о счете не применима", + "连接超时，无法获取对话。": "Тайм-аут подключения, не удалось получить диалог.", + "选择LoRA模型": "Выберите модель LoRA", + "选择Prompt模板集合文件": "Выберите файл с набором шаблонов Prompt", + "选择回复语言（针对搜索&索引功能）": "Выберите язык ответа (для функций поиска и индексации)", + "选择数据集": "Выберите набор данных", + "选择模型": "Выберите модель", + "重命名该对话": "Переименовать этот диалог", + "重新生成": "Пересоздать", + "高级": "Расширенные настройки", + "，本次对话累计消耗了 ": ", Общая стоимость этого диалога составляет ", + "💾 保存对话": "💾 Сохранить диалог", + "📝 导出为 Markdown": "📝 Экспортировать в Markdown", + "🔄 切换API地址": "🔄 Переключить адрес API", + "🔄 刷新": "🔄 Обновить", + "🔄 检查更新...": "🔄 Проверить обновления...", + "🔄 设置代理地址": "🔄 Установить адрес прокси", + "🔄 重新生成": "🔄 Пересоздать", + "🔙 恢复默认网络设置": "🔙 Восстановить настройки сети по умолчанию", + "🗑️ 删除最新对话": "🗑️ Удалить последний диалог", + "🗑️ 删除最旧对话": "🗑️ Удалить старейший диалог", + "🧹 新的对话": "🧹 Новый диалог", + "正在获取IP地址信息，请稍候...": "Получение информации об IP-адресе, пожалуйста, подождите...", +} \ No newline at end of file diff --git a/locale/sv_SE.json b/locale/sv_SE.json new file mode 100644 index 0000000000000000000000000000000000000000..5006daf4299fc883a5f16b712cb4cb42a5d07462 --- /dev/null +++ b/locale/sv_SE.json @@ -0,0 +1,144 @@ +{ + " 吗？": " ?", + "# ⚠️ 务必谨慎更改 ⚠️": "# ⚠️ Var försiktig med ändringar. ⚠️", + "**发送消息** 或 **提交key** 以显示额度": "**Skicka meddelande** eller **Skicka in nyckel** för att visa kredit", + "**本月使用金额** ": "**Månadens användning** ", + "**获取API使用情况失败**": "**Misslyckades med att hämta API-användning**", + "**获取API使用情况失败**，sensitive_id错误或已过期": "**Misslyckades med att hämta API-användning**, felaktig eller utgången sensitive_id", + "**获取API使用情况失败**，需在填写`config.json`中正确填写sensitive_id": "**Misslyckades med att hämta API-användning**, korrekt sensitive_id behövs i `config.json`", + "API key为空，请检查是否输入正确。": "API-nyckeln är tom, kontrollera om den är korrekt inmatad.", + "API密钥更改为了": "API-nyckeln har ändrats till", + "JSON解析错误,收到的内容: ": "JSON-tolkningsfel, mottaget innehåll: ", + "SSL错误，无法获取对话。": "SSL-fel, kunde inte hämta dialogen.", + "Token 计数: ": "Tokenräkning: ", + "☹️发生了错误：": "☹️Fel: ", + "⚠️ 为保证API-Key安全，请在配置文件`config.json`中修改网络设置": "⚠️ För att säkerställa säkerheten för API-nyckeln, vänligen ändra nätverksinställningarna i konfigurationsfilen `config.json`.", + "。你仍然可以使用聊天功能。": ". Du kan fortfarande använda chattfunktionen.", + "上传": "Ladda upp", + "上传了": "Uppladdad", + "上传到 OpenAI 后自动填充": "Automatiskt ifylld efter uppladdning till OpenAI", + "上传到OpenAI": "Ladda upp till OpenAI", + "上传文件": "ladda upp fil", + "仅供查看": "Endast för visning", + "从Prompt模板中加载": "Ladda från Prompt-mall", + "从列表中加载对话": "Ladda dialog från lista", + "代理地址": "Proxyadress", + "代理错误，无法获取对话。": "Proxyfel, kunde inte hämta dialogen.", + "你没有权限访问 GPT4，[进一步了解](https://github.com/GaiZhenbiao/ChuanhuChatGPT/issues/843)": "Du har inte behörighet att komma åt GPT-4, [läs mer](https://github.com/GaiZhenbiao/ChuanhuChatGPT/issues/843)", + "你没有选择任何对话历史": "Du har inte valt någon konversationshistorik.", + "你真的要删除 ": "Är du säker på att du vill ta bort ", + "使用在线搜索": "Använd online-sökning", + "停止符，用英文逗号隔开...": "Skriv in stopptecken här, separerade med kommatecken...", + "关于": "om", + "准备数据集": "Förbered dataset", + "切换亮暗色主题": "Byt ljus/mörk tema", + "删除对话历史成功": "Raderade konversationens historik.", + "删除这轮问答": "Ta bort denna omgång av Q&A", + "刷新状态": "Uppdatera status", + "剩余配额不足，[进一步了解](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/%E5%B8%B8%E8%A7%81%E9%97%AE%E9%A2%98#you-exceeded-your-current-quota-please-check-your-plan-and-billing-details)": "Återstående kvot är otillräcklig, [läs mer](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/%C3%84mnen)", + "加载Prompt模板": "Ladda Prompt-mall", + "单轮对话": "Enkel dialog", + "历史记录（JSON）": "Historikfil (JSON)", + "参数": "Parametrar", + "双栏pdf": "Två-kolumns pdf", + "取消": "Avbryt", + "取消所有任务": "Avbryt alla uppgifter", + "可选，用于区分不同的模型": "Valfritt, används för att särskilja olika modeller", + "启用的工具：": "Aktiverade verktyg: ", + "在工具箱中管理知识库文件": "hantera kunskapsbankfiler i verktygslådan", + "在线搜索": "onlinesökning", + "在这里输入": "Skriv in här", + "在这里输入System Prompt...": "Skriv in System Prompt här...", + "多账号模式已开启，无需输入key，可直接开始对话": "Flerkontoläge är aktiverat, ingen nyckel behövs, du kan starta dialogen direkt", + "好": "OK", + "实时传输回答": "Strömmande utdata", + "对话": "konversation", + "对话历史": "Dialoghistorik", + "对话历史记录": "Dialoghistorik", + "对话命名方式": "Dialognamn", + "导出为 Markdown": "Exportera som Markdown", + "川虎Chat": "Chuanhu Chat", + "川虎Chat 🚀": "Chuanhu Chat 🚀", + "小原同学": "Prinvest Mate", + "小原同学 🚀": "Prinvest Mate 🚀", + "工具箱": "verktygslåda", + "已经被删除啦": "Har raderats.", + "开始实时传输回答……": "Börjar strömma utdata...", + "开始训练": "Börja träning", + "微调": "Finjustering", + "总结": "Sammanfatta", + "总结完成": "Slutfört sammanfattningen.", + "您使用的就是最新版！": "Du använder den senaste versionen!", + "您的IP区域：": "Din IP-region: ", + "您的IP区域：未知。": "Din IP-region: Okänd.", + "拓展": "utvidgning", + "搜索（支持正则）...": "Sök (stöd för reguljära uttryck)...", + "数据集预览": "Datasetförhandsvisning", + "文件ID": "Fil-ID", + "新对话 ": "Ny dialog ", + "新建对话保留Prompt": "Skapa ny konversation med bevarad Prompt", + "暂时未知": "Okänd", + "更新": "Uppdatera", + "更新失败，请尝试[手动更新](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#手动更新)": "Uppdateringen misslyckades, prova att [uppdatera manuellt](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#手动更新)", + "更新成功，请重启本程序": "Uppdaterat framgångsrikt, starta om programmet", + "未命名对话历史记录": "Onämnd Dialoghistorik", + "未设置代理...": "Inte inställd proxy...", + "本月使用金额": "Månadens användning", + "查看[使用介绍](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#微调-gpt-35)": "Se [användarguiden](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#微调-gpt-35) för mer information", + "根据日期时间": "Enligt datum och tid", + "模型": "Modell", + "模型名称后缀": "Modellnamnstillägg", + "模型自动总结（消耗tokens）": "Modellens automatiska sammanfattning (förbrukar tokens)", + "模型设置为了：": "Modellen är inställd på: ", + "正在尝试更新...": "Försöker uppdatera...", + "添加训练好的模型到模型列表": "Lägg till tränad modell i modellistan", + "状态": "Status", + "生成内容总结中……": "Genererar innehållssammanfattning...", + "用于定位滥用行为": "Används för att lokalisera missbruk", + "用户标识符": "Användar-ID", + "由Bilibili [土川虎虎虎](https://space.bilibili.com/29125536)、[明昭MZhao](https://space.bilibili.com/24807452) 和 [Keldos](https://github.com/Keldos-Li) 开发
访问川虎Chat的 [GitHub项目](https://github.com/GaiZhenbiao/ChuanhuChatGPT) 下载最新版脚本": "Utvecklad av Bilibili [土川虎虎虎](https://space.bilibili.com/29125536), [明昭MZhao](https://space.bilibili.com/24807452) och [Keldos](https://github.com/Keldos-Li)\n\nLadda ner senaste koden från [GitHub](https://github.com/GaiZhenbiao/ChuanhuChatGPT)", + "知识库": "kunskapsbank", + "知识库文件": "kunskapsbankfil", + "第一条提问": "Första frågan", + "索引构建完成": "Indexet har blivit byggt färdigt.", + "网络": "nätverksparametrar", + "获取API使用情况失败:": "Misslyckades med att hämta API-användning:", + "获取IP地理位置失败。原因：": "Misslyckades med att hämta IP-plats. Orsak: ", + "获取对话时发生错误，请查看后台日志": "Ett fel uppstod när dialogen hämtades, kontrollera bakgrundsloggen", + "训练": "träning", + "训练状态": "Träningsstatus", + "训练轮数（Epochs）": "Träningsomgångar (Epochs)", + "设置": "inställningar", + "设置保存文件名": "Ställ in sparfilnamn", + "设置文件名: 默认为.json，可选为.md": "Ställ in filnamn: standard är .json, valfritt är .md", + "识别公式": "Formel OCR", + "详情": "Detaljer", + "请查看 config_example.json，配置 Azure OpenAI": "Vänligen granska config_example.json för att konfigurera Azure OpenAI", + "请检查网络连接，或者API-Key是否有效。": "Kontrollera nätverksanslutningen eller om API-nyckeln är giltig.", + "请输入对话内容。": "Ange dialoginnehåll.", + "请输入有效的文件名，不要包含以下特殊字符：": "Ange ett giltigt filnamn, använd inte följande specialtecken: ", + "读取超时，无法获取对话。": "Läsningen tog för lång tid, kunde inte hämta dialogen.", + "账单信息不适用": "Faktureringsinformation är inte tillämplig", + "连接超时，无法获取对话。": "Anslutningen tog för lång tid, kunde inte hämta dialogen.", + "选择LoRA模型": "Välj LoRA Modell", + "选择Prompt模板集合文件": "Välj Prompt-mall Samlingsfil", + "选择回复语言（针对搜索&索引功能）": "Välj svarspråk (för sök- och indexfunktion)", + "选择数据集": "Välj dataset", + "选择模型": "Välj Modell", + "重命名该对话": "Byt namn på dialogen", + "重新生成": "Återgenerera", + "高级": "Avancerat", + "，本次对话累计消耗了 ": ", Total kostnad för denna dialog är ", + "💾 保存对话": "💾 Spara Dialog", + "📝 导出为 Markdown": "📝 Exportera som Markdown", + "🔄 切换API地址": "🔄 Byt API-adress", + "🔄 刷新": "🔄 Uppdatera", + "🔄 检查更新...": "🔄 Sök efter uppdateringar...", + "🔄 设置代理地址": "🔄 Ställ in Proxyadress", + "🔄 重新生成": "🔄 Regenerera", + "🔙 恢复默认网络设置": "🔙 Återställ standardnätverksinställningar+", + "🗑️ 删除最新对话": "🗑️ Ta bort senaste dialogen", + "🗑️ 删除最旧对话": "🗑️ Ta bort äldsta dialogen", + "🧹 新的对话": "🧹 Ny Dialog", + "正在获取IP地址信息，请稍候...": "Hämtar IP-adressinformation, vänta...", +} \ No newline at end of file diff --git a/locale/vi_VN.json b/locale/vi_VN.json new file mode 100644 index 0000000000000000000000000000000000000000..be2e2a82768be766e9572fa02ac2c20f358da1dd --- /dev/null +++ b/locale/vi_VN.json @@ -0,0 +1,144 @@ +{ + " 吗？": " ?", + "# ⚠️ 务必谨慎更改 ⚠️": "# ⚠️ Lưu ý: Thay đổi yêu cầu cẩn thận. ⚠️", + "**发送消息** 或 **提交key** 以显示额度": "**Gửi tin nhắn** hoặc **Gửi khóa(key)** để hiển thị số dư", + "**本月使用金额** ": "**Số tiền sử dụng trong tháng** ", + "**获取API使用情况失败**": "**Lỗi khi lấy thông tin sử dụng API**", + "**获取API使用情况失败**，sensitive_id错误或已过期": "**Lỗi khi lấy thông tin sử dụng API**, sensitive_id sai hoặc đã hết hạn", + "**获取API使用情况失败**，需在填写`config.json`中正确填写sensitive_id": "**Lỗi khi lấy thông tin sử dụng API**, cần điền đúng sensitive_id trong tệp `config.json`", + "API key为空，请检查是否输入正确。": "Khóa API trống, vui lòng kiểm tra xem đã nhập đúng chưa.", + "API密钥更改为了": "Khóa API đã được thay đổi thành", + "JSON解析错误,收到的内容: ": "Lỗi phân tích JSON, nội dung nhận được: ", + "SSL错误，无法获取对话。": "Lỗi SSL, không thể nhận cuộc trò chuyện.", + "Token 计数: ": "Số lượng Token: ", + "☹️发生了错误：": "☹️Lỗi: ", + "⚠️ 为保证API-Key安全，请在配置文件`config.json`中修改网络设置": "⚠️ Để đảm bảo an toàn cho API-Key, vui lòng chỉnh sửa cài đặt mạng trong tệp cấu hình `config.json`.", + "。你仍然可以使用聊天功能。": ". Bạn vẫn có thể sử dụng chức năng trò chuyện.", + "上传": "Tải lên", + "上传了": "Tải lên thành công.", + "上传到 OpenAI 后自动填充": "Tự động điền sau khi tải lên OpenAI", + "上传到OpenAI": "Tải lên OpenAI", + "上传文件": "Tải lên tệp", + "仅供查看": "Chỉ xem", + "从Prompt模板中加载": "Tải từ mẫu Prompt", + "从列表中加载对话": "Tải cuộc trò chuyện từ danh sách", + "代理地址": "Địa chỉ proxy", + "代理错误，无法获取对话。": "Lỗi proxy, không thể nhận cuộc trò chuyện.", + "你没有权限访问 GPT4，[进一步了解](https://github.com/GaiZhenbiao/ChuanhuChatGPT/issues/843)": "Bạn không có quyền truy cập GPT-4, [tìm hiểu thêm](https://github.com/GaiZhenbiao/ChuanhuChatGPT/issues/843)", + "你没有选择任何对话历史": "Bạn chưa chọn bất kỳ lịch sử trò chuyện nào.", + "你真的要删除 ": "Bạn có chắc chắn muốn xóa ", + "使用在线搜索": "Sử dụng tìm kiếm trực tuyến", + "停止符，用英文逗号隔开...": "Nhập dấu dừng, cách nhau bằng dấu phẩy...", + "关于": "Về", + "准备数据集": "Chuẩn bị tập dữ liệu", + "切换亮暗色主题": "Chuyển đổi chủ đề sáng/tối", + "删除对话历史成功": "Xóa lịch sử cuộc trò chuyện thành công.", + "删除这轮问答": "Xóa cuộc trò chuyện này", + "刷新状态": "Làm mới tình trạng", + "剩余配额不足，[进一步了解](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/%E5%B8%B8%E8%A7%81%E9%97%AE%E9%A2%98#you-exceeded-your-current-quota-please-check-your-plan-and-billing-details)": "剩余配额 không đủ, [Nhấn vào đây để biết thêm](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/%E5%B8%B8%E8%A7%81%E9%97%AE%E9%A2%98#you-exceeded-your-current-quota-please-check-your-plan-and-billing-details)", + "加载Prompt模板": "Tải mẫu Prompt", + "单轮对话": "Cuộc trò chuyện một lượt", + "历史记录（JSON）": "Tệp lịch sử (JSON)", + "参数": "Tham số", + "双栏pdf": "PDF hai cột", + "取消": "Hủy", + "取消所有任务": "Hủy tất cả các nhiệm vụ", + "可选，用于区分不同的模型": "Tùy chọn, sử dụng để phân biệt các mô hình khác nhau", + "启用的工具：": "Công cụ đã bật: ", + "在工具箱中管理知识库文件": "Quản lý tệp cơ sở kiến thức trong hộp công cụ", + "在线搜索": "Tìm kiếm trực tuyến", + "在这里输入": "Nhập vào đây", + "在这里输入System Prompt...": "Nhập System Prompt ở đây...", + "多账号模式已开启，无需输入key，可直接开始对话": "Chế độ nhiều tài khoản đã được bật, không cần nhập key, bạn có thể bắt đầu cuộc trò chuyện trực tiếp", + "好": "OK", + "实时传输回答": "Truyền đầu ra trực tiếp", + "对话": "Cuộc trò chuyện", + "对话历史": "Lịch sử cuộc trò chuyện", + "对话历史记录": "Lịch sử Cuộc trò chuyện", + "对话命名方式": "Phương thức đặt tên lịch sử trò chuyện", + "导出为 Markdown": "Xuất ra Markdown", + "川虎Chat": "Chuanhu Chat", + "川虎Chat 🚀": "Chuanhu Chat 🚀", + "小原同学": "Prinvest Mate", + "小原同学 🚀": "Prinvest Mate 🚀", + "工具箱": "Hộp công cụ", + "已经被删除啦": "Đã bị xóa rồi.", + "开始实时传输回答……": "Bắt đầu truyền đầu ra trực tiếp...", + "开始训练": "Bắt đầu đào tạo", + "微调": "Feeling-tuning", + "总结": "Tóm tắt", + "总结完成": "Hoàn thành tóm tắt", + "您使用的就是最新版！": "Bạn đang sử dụng phiên bản mới nhất!", + "您的IP区域：": "Khu vực IP của bạn: ", + "您的IP区域：未知。": "Khu vực IP của bạn: Không xác định.", + "拓展": "Mở rộng", + "搜索（支持正则）...": "Tìm kiếm (hỗ trợ regex)...", + "数据集预览": "Xem trước tập dữ liệu", + "文件ID": "ID Tệp", + "新对话 ": "Cuộc trò chuyện mới ", + "新建对话保留Prompt": "Tạo Cuộc trò chuyện mới và giữ Prompt nguyên vẹn", + "暂时未知": "Tạm thời chưa xác định", + "更新": "Cập nhật", + "更新失败，请尝试[手动更新](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#手动更新)": "Cập nhật thất bại, vui lòng thử [cập nhật thủ công](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#手动更新)", + "更新成功，请重启本程序": "Cập nhật thành công, vui lòng khởi động lại chương trình này", + "未命名对话历史记录": "Lịch sử Cuộc trò chuyện không đặt tên", + "未设置代理...": "Không có proxy...", + "本月使用金额": "Số tiền sử dụng trong tháng", + "查看[使用介绍](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#微调-gpt-35)": "Xem [hướng dẫn sử dụng](https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki/使用教程#微调-gpt-35) để biết thêm chi tiết", + "根据日期时间": "Theo ngày và giờ", + "模型": "Mô hình", + "模型名称后缀": "Hậu tố Tên Mô hình", + "模型自动总结（消耗tokens）": "Tự động tóm tắt bằng LLM (Tiêu thụ token)", + "模型设置为了：": "Mô hình đã được đặt thành: ", + "正在尝试更新...": "Đang cố gắng cập nhật...", + "添加训练好的模型到模型列表": "Thêm mô hình đã đào tạo vào danh sách mô hình", + "状态": "Tình trạng", + "生成内容总结中……": "Đang tạo tóm tắt nội dung...", + "用于定位滥用行为": "Sử dụng để xác định hành vi lạm dụng", + "用户标识符": "Định danh người dùng", + "由Bilibili [土川虎虎虎](https://space.bilibili.com/29125536)、[明昭MZhao](https://space.bilibili.com/24807452) 和 [Keldos](https://github.com/Keldos-Li) 开发
访问川虎Chat的 [GitHub项目](https://github.com/GaiZhenbiao/ChuanhuChatGPT) 下载最新版脚本": "Phát triển bởi Bilibili [土川虎虎虎](https://space.bilibili.com/29125536), [明昭MZhao](https://space.bilibili.com/24807452) và [Keldos](https://github.com/Keldos-Li)\n\nTải mã nguồn mới nhất từ [GitHub](https://github.com/GaiZhenbiao/ChuanhuChatGPT)", + "知识库": "Cơ sở kiến thức", + "知识库文件": "Tệp cơ sở kiến thức", + "第一条提问": "Theo câu hỏi đầu tiên", + "索引构建完成": "Xây dựng chỉ mục hoàn tất", + "网络": "Mạng", + "获取API使用情况失败:": "Lỗi khi lấy thông tin sử dụng API:", + "获取IP地理位置失败。原因：": "Không thể lấy vị trí địa lý của IP. Nguyên nhân: ", + "获取对话时发生错误，请查看后台日志": "Xảy ra lỗi khi nhận cuộc trò chuyện, kiểm tra nhật ký nền", + "训练": "Đào tạo", + "训练状态": "Tình trạng đào tạo", + "训练轮数（Epochs）": "Số lượt đào tạo (Epochs)", + "设置": "Cài đặt", + "设置保存文件名": "Đặt tên tệp lưu", + "设置文件名: 默认为.json，可选为.md": "Đặt tên tệp: mặc định là .json, tùy chọn là .md", + "识别公式": "Nhận dạng công thức", + "详情": "Chi tiết", + "请查看 config_example.json，配置 Azure OpenAI": "Vui lòng xem tệp config_example.json để cấu hình Azure OpenAI", + "请检查网络连接，或者API-Key是否有效。": "Vui lòng kiểm tra kết nối mạng hoặc xem xét tính hợp lệ của API-Key.", + "请输入对话内容。": "Nhập nội dung cuộc trò chuyện.", + "请输入有效的文件名，不要包含以下特殊字符：": "Vui lòng nhập tên tệp hợp lệ, không chứa các ký tự đặc biệt sau: ", + "读取超时，无法获取对话。": "Hết thời gian đọc, không thể nhận cuộc trò chuyện.", + "账单信息不适用": "Thông tin thanh toán không áp dụng", + "连接超时，无法获取对话。": "Hết thời gian kết nối, không thể nhận cuộc trò chuyện.", + "选择LoRA模型": "Chọn Mô hình LoRA", + "选择Prompt模板集合文件": "Chọn Tệp bộ sưu tập mẫu Prompt", + "选择回复语言（针对搜索&索引功能）": "Chọn ngôn ngữ phản hồi (đối với chức năng tìm kiếm & chỉ mục)", + "选择数据集": "Chọn tập dữ liệu", + "选择模型": "Chọn Mô hình", + "重命名该对话": "Đổi tên cuộc trò chuyện này", + "重新生成": "Tạo lại", + "高级": "Nâng cao", + "，本次对话累计消耗了 ": ", Tổng cộng chi phí cho cuộc trò chuyện này là ", + "💾 保存对话": "💾 Lưu Cuộc trò chuyện", + "📝 导出为 Markdown": "📝 Xuất ra dưới dạng Markdown", + "🔄 切换API地址": "🔄 Chuyển đổi Địa chỉ API", + "🔄 刷新": "🔄 Làm mới", + "🔄 检查更新...": "🔄 Kiểm tra cập nhật...", + "🔄 设置代理地址": "🔄 Đặt Địa chỉ Proxy", + "🔄 重新生成": "🔄 Tạo lại", + "🔙 恢复默认网络设置": "🔙 Khôi phục cài đặt mạng mặc định", + "🗑️ 删除最新对话": "🗑️ Xóa cuộc trò chuyện mới nhất", + "🗑️ 删除最旧对话": "🗑️ Xóa cuộc trò chuyện cũ nhất", + "🧹 新的对话": "🧹 Cuộc trò chuyện mới", + "正在获取IP地址信息，请稍候...": "Đang lấy thông tin địa chỉ IP, vui lòng đợi...", +} \ No newline at end of file diff --git a/locale/zh_CN.json b/locale/zh_CN.json new file mode 100644 index 0000000000000000000000000000000000000000..9e26dfeeb6e641a33dae4961196235bdb965b21b --- /dev/null +++ b/locale/zh_CN.json @@ -0,0 +1 @@ +{} \ No newline at end of file diff --git a/modules/__init__.py b/modules/__init__.py new file mode 100644 index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 diff --git a/modules/config.py b/modules/config.py new file mode 100644 index 0000000000000000000000000000000000000000..1b18fd64ce1f0b7fcf5201c10aa67cfb5fd77314 --- /dev/null +++ b/modules/config.py @@ -0,0 +1,308 @@ +from collections import defaultdict +from contextlib import contextmanager +import os +import logging +import sys +import commentjson as json + +from . import shared +from . import presets + + +__all__ = [ + "my_api_key", + "sensitive_id", + "authflag", + "auth_list", + "dockerflag", + "retrieve_proxy", + "advance_docs", + "update_doc_config", + "usage_limit", + "multi_api_key", + "server_name", + "server_port", + "share", + "check_update", + "latex_delimiters_set", + "hide_history_when_not_logged_in", + "default_chuanhu_assistant_model", + "show_api_billing", + "chat_name_method_index", + "HIDE_MY_KEY", +] + +# 添加一个统一的config文件，避免文件过多造成的疑惑（优先级最低） +# 同时，也可以为后续支持自定义功能提供config的帮助 +if os.path.exists("config.json"): + with open("config.json", "r", encoding='utf-8') as f: + config = json.load(f) +else: + config = {} + + +def load_config_to_environ(key_list): + global config + for key in key_list: + if key in config: + os.environ[key.upper()] = os.environ.get(key.upper(), config[key]) + +hide_history_when_not_logged_in = config.get( + "hide_history_when_not_logged_in", False) +check_update = config.get("check_update", True) +show_api_billing = config.get("show_api_billing", False) +show_api_billing = bool(os.environ.get("SHOW_API_BILLING", show_api_billing)) +chat_name_method_index = config.get("chat_name_method_index", 2) + +if os.path.exists("api_key.txt"): + logging.info("检测到api_key.txt文件，正在进行迁移...") + with open("api_key.txt", "r", encoding="utf-8") as f: + config["openai_api_key"] = f.read().strip() + os.rename("api_key.txt", "api_key(deprecated).txt") + with open("config.json", "w", encoding='utf-8') as f: + json.dump(config, f, indent=4, ensure_ascii=False) + +if os.path.exists("auth.json"): + logging.info("检测到auth.json文件，正在进行迁移...") + auth_list = [] + with open("auth.json", "r", encoding='utf-8') as f: + auth = json.load(f) + for _ in auth: + if auth[_]["username"] and auth[_]["password"]: + auth_list.append((auth[_]["username"], auth[_]["password"])) + else: + logging.error("请检查auth.json文件中的用户名和密码！") + sys.exit(1) + config["users"] = auth_list + os.rename("auth.json", "auth(deprecated).json") + with open("config.json", "w", encoding='utf-8') as f: + json.dump(config, f, indent=4, ensure_ascii=False) + +# 处理docker if we are running in Docker +dockerflag = config.get("dockerflag", False) +if os.environ.get("dockerrun") == "yes": + dockerflag = True + +# 处理 api-key 以及 允许的用户列表 +my_api_key = config.get("openai_api_key", "") +my_api_key = os.environ.get("OPENAI_API_KEY", my_api_key) +os.environ["OPENAI_API_KEY"] = my_api_key +os.environ["OPENAI_EMBEDDING_API_KEY"] = my_api_key + +if config.get("legacy_api_usage", False): + sensitive_id = my_api_key +else: + sensitive_id = config.get("sensitive_id", "") + sensitive_id = os.environ.get("SENSITIVE_ID", sensitive_id) + +if "available_models" in config: + presets.MODELS = config["available_models"] + logging.info(f"已设置可用模型：{config['available_models']}") + +# 模型配置 +if "extra_models" in config: + presets.MODELS.extend(config["extra_models"]) + logging.info(f"已添加额外的模型：{config['extra_models']}") + +HIDE_MY_KEY = config.get("hide_my_key", False) + +google_palm_api_key = config.get("google_palm_api_key", "") +google_palm_api_key = os.environ.get( + "GOOGLE_PALM_API_KEY", google_palm_api_key) +os.environ["GOOGLE_PALM_API_KEY"] = google_palm_api_key + +xmchat_api_key = config.get("xmchat_api_key", "") +os.environ["XMCHAT_API_KEY"] = xmchat_api_key + +minimax_api_key = config.get("minimax_api_key", "") +os.environ["MINIMAX_API_KEY"] = minimax_api_key +minimax_group_id = config.get("minimax_group_id", "") +os.environ["MINIMAX_GROUP_ID"] = minimax_group_id + +midjourney_proxy_api_base = config.get("midjourney_proxy_api_base", "") +os.environ["MIDJOURNEY_PROXY_API_BASE"] = midjourney_proxy_api_base +midjourney_proxy_api_secret = config.get("midjourney_proxy_api_secret", "") +os.environ["MIDJOURNEY_PROXY_API_SECRET"] = midjourney_proxy_api_secret +midjourney_discord_proxy_url = config.get("midjourney_discord_proxy_url", "") +os.environ["MIDJOURNEY_DISCORD_PROXY_URL"] = midjourney_discord_proxy_url +midjourney_temp_folder = config.get("midjourney_temp_folder", "") +os.environ["MIDJOURNEY_TEMP_FOLDER"] = midjourney_temp_folder + +spark_api_key = config.get("spark_api_key", "") +os.environ["SPARK_API_KEY"] = spark_api_key +spark_appid = config.get("spark_appid", "") +os.environ["SPARK_APPID"] = spark_appid +spark_api_secret = config.get("spark_api_secret", "") +os.environ["SPARK_API_SECRET"] = spark_api_secret + +claude_api_secret = config.get("claude_api_secret", "") +os.environ["CLAUDE_API_SECRET"] = claude_api_secret + +ernie_api_key = config.get("ernie_api_key", "") +os.environ["ERNIE_APIKEY"] = ernie_api_key +ernie_secret_key = config.get("ernie_secret_key", "") +os.environ["ERNIE_SECRETKEY"] = ernie_secret_key + +load_config_to_environ(["openai_api_type", "azure_openai_api_key", "azure_openai_api_base_url", + "azure_openai_api_version", "azure_deployment_name", "azure_embedding_deployment_name", "azure_embedding_model_name"]) + + +usage_limit = os.environ.get("USAGE_LIMIT", config.get("usage_limit", 120)) + +# 多账户机制 +multi_api_key = config.get("multi_api_key", False) # 是否开启多账户机制 +if multi_api_key: + api_key_list = config.get("api_key_list", []) + if len(api_key_list) == 0: + logging.error("多账号模式已开启，但api_key_list为空，请检查config.json") + sys.exit(1) + shared.state.set_api_key_queue(api_key_list) + +auth_list = config.get("users", []) # 实际上是使用者的列表 +authflag = len(auth_list) > 0 # 是否开启认证的状态值，改为判断auth_list长度 + +# 处理自定义的api_host，优先读环境变量的配置，如果存在则自动装配 +api_host = os.environ.get( + "OPENAI_API_BASE", config.get("openai_api_base", None)) +if api_host is not None: + shared.state.set_api_host(api_host) + os.environ["OPENAI_API_BASE"] = f"{api_host}" + logging.info(f"OpenAI API Base set to: {os.environ['OPENAI_API_BASE']}") + +default_chuanhu_assistant_model = config.get( + "default_chuanhu_assistant_model", "gpt-3.5-turbo") +for x in ["GOOGLE_CSE_ID", "GOOGLE_API_KEY", "WOLFRAM_ALPHA_APPID", "SERPAPI_API_KEY"]: + if config.get(x, None) is not None: + os.environ[x] = config[x] + + +@contextmanager +def retrieve_openai_api(api_key=None): + old_api_key = os.environ.get("OPENAI_API_KEY", "") + if api_key is None: + os.environ["OPENAI_API_KEY"] = my_api_key + yield my_api_key + else: + os.environ["OPENAI_API_KEY"] = api_key + yield api_key + os.environ["OPENAI_API_KEY"] = old_api_key + + + +# 处理代理： +http_proxy = os.environ.get("HTTP_PROXY", "") +https_proxy = os.environ.get("HTTPS_PROXY", "") +http_proxy = config.get("http_proxy", http_proxy) +https_proxy = config.get("https_proxy", https_proxy) + +# 重置系统变量，在不需要设置的时候不设置环境变量，以免引起全局代理报错 +os.environ["HTTP_PROXY"] = "" +os.environ["HTTPS_PROXY"] = "" + +local_embedding = config.get("local_embedding", False) # 是否使用本地embedding + + +@contextmanager +def retrieve_proxy(proxy=None): + """ + 1, 如果proxy = NONE，设置环境变量，并返回最新设置的代理 + 2，如果proxy ！= NONE，更新当前的代理配置，但是不更新环境变量 + """ + global http_proxy, https_proxy + if proxy is not None: + http_proxy = proxy + https_proxy = proxy + yield http_proxy, https_proxy + else: + old_var = os.environ["HTTP_PROXY"], os.environ["HTTPS_PROXY"] + os.environ["HTTP_PROXY"] = http_proxy + os.environ["HTTPS_PROXY"] = https_proxy + yield http_proxy, https_proxy # return new proxy + + # return old proxy + os.environ["HTTP_PROXY"], os.environ["HTTPS_PROXY"] = old_var + + +# 处理latex options +user_latex_option = config.get("latex_option", "default") +if user_latex_option == "default": + latex_delimiters_set = [ + {"left": "$$", "right": "$$", "display": True}, + {"left": "$", "right": "$", "display": False}, + {"left": "\\(", "right": "\\)", "display": False}, + {"left": "\\[", "right": "\\]", "display": True}, + ] +elif user_latex_option == "strict": + latex_delimiters_set = [ + {"left": "$$", "right": "$$", "display": True}, + {"left": "\\(", "right": "\\)", "display": False}, + {"left": "\\[", "right": "\\]", "display": True}, + ] +elif user_latex_option == "all": + latex_delimiters_set = [ + {"left": "$$", "right": "$$", "display": True}, + {"left": "$", "right": "$", "display": False}, + {"left": "\\(", "right": "\\)", "display": False}, + {"left": "\\[", "right": "\\]", "display": True}, + {"left": "\\begin{equation}", "right": "\\end{equation}", "display": True}, + {"left": "\\begin{align}", "right": "\\end{align}", "display": True}, + {"left": "\\begin{alignat}", "right": "\\end{alignat}", "display": True}, + {"left": "\\begin{gather}", "right": "\\end{gather}", "display": True}, + {"left": "\\begin{CD}", "right": "\\end{CD}", "display": True}, + ] +elif user_latex_option == "disabled": + latex_delimiters_set = [] +else: + latex_delimiters_set = [ + {"left": "$$", "right": "$$", "display": True}, + {"left": "$", "right": "$", "display": False}, + {"left": "\\(", "right": "\\)", "display": False}, + {"left": "\\[", "right": "\\]", "display": True}, + ] + +# 处理advance docs +advance_docs = defaultdict(lambda: defaultdict(dict)) +advance_docs.update(config.get("advance_docs", {})) + + +def update_doc_config(two_column_pdf): + global advance_docs + advance_docs["pdf"]["two_column"] = two_column_pdf + + logging.info(f"更新后的文件参数为：{advance_docs}") + + +# 处理gradio.launch参数 +server_name = config.get("server_name", None) +server_port = config.get("server_port", None) +if server_name is None: + if dockerflag: + server_name = "0.0.0.0" + else: + server_name = "127.0.0.1" +if server_port is None: + if dockerflag: + server_port = 7860 + +assert server_port is None or type(server_port) == int, "要求port设置为int类型" + +# 设置默认model +default_model = config.get("default_model", "") +try: + presets.DEFAULT_MODEL = presets.MODELS.index(default_model) +except ValueError: + pass + +share = config.get("share", False) + +# avatar +bot_avatar = config.get("bot_avatar", "default") +user_avatar = config.get("user_avatar", "default") +if bot_avatar == "" or bot_avatar == "none" or bot_avatar is None: + bot_avatar = None +elif bot_avatar == "default": + bot_avatar = "web_assets/chatbot.png" +if user_avatar == "" or user_avatar == "none" or user_avatar is None: + user_avatar = None +elif user_avatar == "default": + user_avatar = "web_assets/user.png" diff --git a/modules/index_func.py b/modules/index_func.py new file mode 100644 index 0000000000000000000000000000000000000000..d19557053728e7946583078d10404943d1a7b32b --- /dev/null +++ b/modules/index_func.py @@ -0,0 +1,139 @@ +import os +import logging + +import hashlib +import PyPDF2 +from tqdm import tqdm + +from modules.presets import * +from modules.utils import * +from modules.config import local_embedding + + +def get_documents(file_src): + from langchain.schema import Document + from langchain.text_splitter import TokenTextSplitter + text_splitter = TokenTextSplitter(chunk_size=500, chunk_overlap=30) + + documents = [] + logging.debug("Loading documents...") + logging.debug(f"file_src: {file_src}") + for file in file_src: + filepath = file.name + filename = os.path.basename(filepath) + file_type = os.path.splitext(filename)[1] + logging.info(f"loading file: {filename}") + texts = None + try: + if file_type == ".pdf": + logging.debug("Loading PDF...") + try: + from modules.pdf_func import parse_pdf + from modules.config import advance_docs + + two_column = advance_docs["pdf"].get("two_column", False) + pdftext = parse_pdf(filepath, two_column).text + except: + pdftext = "" + with open(filepath, "rb") as pdfFileObj: + pdfReader = PyPDF2.PdfReader(pdfFileObj) + for page in tqdm(pdfReader.pages): + pdftext += page.extract_text() + texts = [Document(page_content=pdftext, + metadata={"source": filepath})] + elif file_type == ".docx": + logging.debug("Loading Word...") + from langchain.document_loaders import UnstructuredWordDocumentLoader + loader = UnstructuredWordDocumentLoader(filepath) + texts = loader.load() + elif file_type == ".pptx": + logging.debug("Loading PowerPoint...") + from langchain.document_loaders import UnstructuredPowerPointLoader + loader = UnstructuredPowerPointLoader(filepath) + texts = loader.load() + elif file_type == ".epub": + logging.debug("Loading EPUB...") + from langchain.document_loaders import UnstructuredEPubLoader + loader = UnstructuredEPubLoader(filepath) + texts = loader.load() + elif file_type == ".xlsx": + logging.debug("Loading Excel...") + text_list = excel_to_string(filepath) + texts = [] + for elem in text_list: + texts.append(Document(page_content=elem, + metadata={"source": filepath})) + else: + logging.debug("Loading text file...") + from langchain.document_loaders import TextLoader + loader = TextLoader(filepath, "utf8") + texts = loader.load() + except Exception as e: + import traceback + logging.error(f"Error loading file: {filename}") + traceback.print_exc() + + if texts is not None: + texts = text_splitter.split_documents(texts) + documents.extend(texts) + logging.debug("Documents loaded.") + return documents + + +def construct_index( + api_key, + file_src, + max_input_size=4096, + num_outputs=5, + max_chunk_overlap=20, + chunk_size_limit=600, + embedding_limit=None, + separator=" ", + load_from_cache_if_possible=True, +): + from langchain.chat_models import ChatOpenAI + from langchain.vectorstores import FAISS + + if api_key: + os.environ["OPENAI_API_KEY"] = api_key + else: + # 由于一个依赖的愚蠢的设计，这里必须要有一个API KEY + os.environ["OPENAI_API_KEY"] = "sk-xxxxxxx" + chunk_size_limit = None if chunk_size_limit == 0 else chunk_size_limit + embedding_limit = None if embedding_limit == 0 else embedding_limit + separator = " " if separator == "" else separator + + index_name = get_file_hash(file_src) + index_path = f"./index/{index_name}" + if local_embedding: + from langchain.embeddings.huggingface import HuggingFaceEmbeddings + embeddings = HuggingFaceEmbeddings( + model_name="sentence-transformers/distiluse-base-multilingual-cased-v2") + else: + from langchain.embeddings import OpenAIEmbeddings + if os.environ.get("OPENAI_API_TYPE", "openai") == "openai": + embeddings = OpenAIEmbeddings(openai_api_base=os.environ.get( + "OPENAI_API_BASE", None), openai_api_key=os.environ.get("OPENAI_EMBEDDING_API_KEY", api_key)) + else: + embeddings = OpenAIEmbeddings(deployment=os.environ["AZURE_EMBEDDING_DEPLOYMENT_NAME"], openai_api_key=os.environ["AZURE_OPENAI_API_KEY"], + model=os.environ["AZURE_EMBEDDING_MODEL_NAME"], openai_api_base=os.environ["AZURE_OPENAI_API_BASE_URL"], openai_api_type="azure") + if os.path.exists(index_path) and load_from_cache_if_possible: + logging.info("找到了缓存的索引文件，加载中……") + return FAISS.load_local(index_path, embeddings) + else: + try: + documents = get_documents(file_src) + logging.info("构建索引中……") + with retrieve_proxy(): + index = FAISS.from_documents(documents, embeddings) + logging.debug("索引构建完成！") + os.makedirs("./index", exist_ok=True) + index.save_local(index_path) + logging.debug("索引已保存至本地!") + return index + + except Exception as e: + import traceback + logging.error("索引构建失败！%s", e) + traceback.print_exc() + return None diff --git a/modules/models/Azure.py b/modules/models/Azure.py new file mode 100644 index 0000000000000000000000000000000000000000..f6c7adaadd57c6860609889d298981ce5f31146c --- /dev/null +++ b/modules/models/Azure.py @@ -0,0 +1,18 @@ +from langchain.chat_models import AzureChatOpenAI, ChatOpenAI +import os + +from .base_model import Base_Chat_Langchain_Client + +# load_config_to_environ(["azure_openai_api_key", "azure_api_base_url", "azure_openai_api_version", "azure_deployment_name"]) + +class Azure_OpenAI_Client(Base_Chat_Langchain_Client): + def setup_model(self): + # inplement this to setup the model then return it + return AzureChatOpenAI( + openai_api_base=os.environ["AZURE_OPENAI_API_BASE_URL"], + openai_api_version=os.environ["AZURE_OPENAI_API_VERSION"], + deployment_name=os.environ["AZURE_DEPLOYMENT_NAME"], + openai_api_key=os.environ["AZURE_OPENAI_API_KEY"], + openai_api_type="azure", + streaming=True + ) diff --git a/modules/models/ChatGLM.py b/modules/models/ChatGLM.py new file mode 100644 index 0000000000000000000000000000000000000000..da0be6742e8d5469350a3289957cd0ab8c550efc --- /dev/null +++ b/modules/models/ChatGLM.py @@ -0,0 +1,107 @@ +from __future__ import annotations + +import logging +import os +import platform + +import gc +import torch +import colorama + +from ..index_func import * +from ..presets import * +from ..utils import * +from .base_model import BaseLLMModel + + +class ChatGLM_Client(BaseLLMModel): + def __init__(self, model_name, user_name="") -> None: + super().__init__(model_name=model_name, user=user_name) + import torch + from transformers import AutoModel, AutoTokenizer + global CHATGLM_TOKENIZER, CHATGLM_MODEL + self.deinitialize() + if CHATGLM_TOKENIZER is None or CHATGLM_MODEL is None: + system_name = platform.system() + model_path = None + if os.path.exists("models"): + model_dirs = os.listdir("models") + if model_name in model_dirs: + model_path = f"models/{model_name}" + if model_path is not None: + model_source = model_path + else: + model_source = f"THUDM/{model_name}" + CHATGLM_TOKENIZER = AutoTokenizer.from_pretrained( + model_source, trust_remote_code=True + ) + quantified = False + if "int4" in model_name: + quantified = True + model = AutoModel.from_pretrained( + model_source, trust_remote_code=True + ) + if torch.cuda.is_available(): + # run on CUDA + logging.info("CUDA is available, using CUDA") + model = model.half().cuda() + # mps加速还存在一些问题，暂时不使用 + elif system_name == "Darwin" and model_path is not None and not quantified: + logging.info("Running on macOS, using MPS") + # running on macOS and model already downloaded + model = model.half().to("mps") + else: + logging.info("GPU is not available, using CPU") + model = model.float() + model = model.eval() + CHATGLM_MODEL = model + + def _get_glm3_style_input(self): + history = self.history + query = history.pop()["content"] + return history, query + + def _get_glm2_style_input(self): + history = [x["content"] for x in self.history] + query = history.pop() + logging.debug(colorama.Fore.YELLOW + + f"{history}" + colorama.Fore.RESET) + assert ( + len(history) % 2 == 0 + ), f"History should be even length. current history is: {history}" + history = [[history[i], history[i + 1]] + for i in range(0, len(history), 2)] + return history, query + + def _get_glm_style_input(self): + if "glm2" in self.model_name: + return self._get_glm2_style_input() + else: + return self._get_glm3_style_input() + + def get_answer_at_once(self): + history, query = self._get_glm_style_input() + response, _ = CHATGLM_MODEL.chat( + CHATGLM_TOKENIZER, query, history=history) + return response, len(response) + + def get_answer_stream_iter(self): + history, query = self._get_glm_style_input() + for response, history in CHATGLM_MODEL.stream_chat( + CHATGLM_TOKENIZER, + query, + history, + max_length=self.token_upper_limit, + top_p=self.top_p, + temperature=self.temperature, + ): + yield response + + def deinitialize(self): + # 释放显存 + global CHATGLM_MODEL, CHATGLM_TOKENIZER + CHATGLM_MODEL = None + CHATGLM_TOKENIZER = None + gc.collect() + torch.cuda.empty_cache() + logging.info("ChatGLM model deinitialized") diff --git a/modules/models/ChuanhuAgent.py b/modules/models/ChuanhuAgent.py new file mode 100644 index 0000000000000000000000000000000000000000..6a13c75d866fbcf16ac9876d466519fd14798528 --- /dev/null +++ b/modules/models/ChuanhuAgent.py @@ -0,0 +1,232 @@ +from langchain.chains.summarize import load_summarize_chain +from langchain import PromptTemplate, LLMChain +from langchain.chat_models import ChatOpenAI +from langchain.prompts import PromptTemplate +from langchain.text_splitter import TokenTextSplitter +from langchain.embeddings import OpenAIEmbeddings +from langchain.vectorstores import FAISS +from langchain.chains import RetrievalQA +from langchain.agents import load_tools +from langchain.agents import initialize_agent +from langchain.agents import AgentType +from langchain.docstore.document import Document +from langchain.tools import BaseTool, StructuredTool, Tool, tool +from langchain.callbacks.stdout import StdOutCallbackHandler +from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler +from langchain.callbacks.base import BaseCallbackManager +from duckduckgo_search import DDGS +from itertools import islice + +from typing import Any, Dict, List, Optional, Union + +from langchain.callbacks.base import BaseCallbackHandler +from langchain.input import print_text +from langchain.schema import AgentAction, AgentFinish, LLMResult + +from pydantic import BaseModel, Field + +import requests +from bs4 import BeautifulSoup +from threading import Thread, Condition +from collections import deque + +from .base_model import BaseLLMModel, CallbackToIterator, ChuanhuCallbackHandler +from ..config import default_chuanhu_assistant_model +from ..presets import SUMMARIZE_PROMPT, i18n +from ..index_func import construct_index + +from langchain.callbacks import get_openai_callback +import os +import gradio as gr +import logging + +class GoogleSearchInput(BaseModel): + keywords: str = Field(description="keywords to search") + +class WebBrowsingInput(BaseModel): + url: str = Field(description="URL of a webpage") + +class WebAskingInput(BaseModel): + url: str = Field(description="URL of a webpage") + question: str = Field(description="Question that you want to know the answer to, based on the webpage's content.") + + +class ChuanhuAgent_Client(BaseLLMModel): + def __init__(self, model_name, openai_api_key, user_name="") -> None: + super().__init__(model_name=model_name, user=user_name) + self.text_splitter = TokenTextSplitter(chunk_size=500, chunk_overlap=30) + self.api_key = openai_api_key + self.llm = ChatOpenAI(openai_api_key=openai_api_key, temperature=0, model_name=default_chuanhu_assistant_model, openai_api_base=os.environ.get("OPENAI_API_BASE", None)) + self.cheap_llm = ChatOpenAI(openai_api_key=openai_api_key, temperature=0, model_name="gpt-3.5-turbo", openai_api_base=os.environ.get("OPENAI_API_BASE", None)) + PROMPT = PromptTemplate(template=SUMMARIZE_PROMPT, input_variables=["text"]) + self.summarize_chain = load_summarize_chain(self.cheap_llm, chain_type="map_reduce", return_intermediate_steps=True, map_prompt=PROMPT, combine_prompt=PROMPT) + self.index_summary = None + self.index = None + if "Pro" in self.model_name: + tools_to_enable = ["llm-math", "arxiv", "wikipedia"] + # if exists GOOGLE_CSE_ID and GOOGLE_API_KEY, enable google-search-results-json + if os.environ.get("GOOGLE_CSE_ID", None) is not None and os.environ.get("GOOGLE_API_KEY", None) is not None: + tools_to_enable.append("google-search-results-json") + else: + logging.warning("GOOGLE_CSE_ID and/or GOOGLE_API_KEY not found, google-search-results-json is disabled.") + # if exists WOLFRAM_ALPHA_APPID, enable wolfram-alpha + if os.environ.get("WOLFRAM_ALPHA_APPID", None) is not None: + tools_to_enable.append("wolfram-alpha") + else: + logging.warning("WOLFRAM_ALPHA_APPID not found, wolfram-alpha is disabled.") + # if exists SERPAPI_API_KEY, enable serpapi + if os.environ.get("SERPAPI_API_KEY", None) is not None: + tools_to_enable.append("serpapi") + else: + logging.warning("SERPAPI_API_KEY not found, serpapi is disabled.") + self.tools = load_tools(tools_to_enable, llm=self.llm) + else: + self.tools = load_tools(["ddg-search", "llm-math", "arxiv", "wikipedia"], llm=self.llm) + self.tools.append( + Tool.from_function( + func=self.google_search_simple, + name="Google Search JSON", + description="useful when you need to search the web.", + args_schema=GoogleSearchInput + ) + ) + + self.tools.append( + Tool.from_function( + func=self.summary_url, + name="Summary Webpage", + description="useful when you need to know the overall content of a webpage.", + args_schema=WebBrowsingInput + ) + ) + + self.tools.append( + StructuredTool.from_function( + func=self.ask_url, + name="Ask Webpage", + description="useful when you need to ask detailed questions about a webpage.", + args_schema=WebAskingInput + ) + ) + + def google_search_simple(self, query): + results = [] + with DDGS() as ddgs: + ddgs_gen = ddgs.text(query, backend="lite") + for r in islice(ddgs_gen, 10): + results.append({ + "title": r["title"], + "link": r["href"], + "snippet": r["body"] + }) + return str(results) + + def handle_file_upload(self, files, chatbot, language): + """if the model accepts multi modal input, implement this function""" + status = gr.Markdown.update() + if files: + index = construct_index(self.api_key, file_src=files) + assert index is not None, "获取索引失败" + self.index = index + status = i18n("索引构建完成") + # Summarize the document + logging.info(i18n("生成内容总结中……")) + with get_openai_callback() as cb: + os.environ["OPENAI_API_KEY"] = self.api_key + from langchain.chains.summarize import load_summarize_chain + from langchain.prompts import PromptTemplate + from langchain.chat_models import ChatOpenAI + prompt_template = "Write a concise summary of the following:\n\n{text}\n\nCONCISE SUMMARY IN " + language + ":" + PROMPT = PromptTemplate(template=prompt_template, input_variables=["text"]) + llm = ChatOpenAI() + chain = load_summarize_chain(llm, chain_type="map_reduce", return_intermediate_steps=True, map_prompt=PROMPT, combine_prompt=PROMPT) + summary = chain({"input_documents": list(index.docstore.__dict__["_dict"].values())}, return_only_outputs=True)["output_text"] + logging.info(f"Summary: {summary}") + self.index_summary = summary + chatbot.append((f"Uploaded {len(files)} files", summary)) + logging.info(cb) + return gr.Files.update(), chatbot, status + + def query_index(self, query): + if self.index is not None: + retriever = self.index.as_retriever() + qa = RetrievalQA.from_chain_type(llm=self.llm, chain_type="stuff", retriever=retriever) + return qa.run(query) + else: + "Error during query." + + def summary(self, text): + texts = Document(page_content=text) + texts = self.text_splitter.split_documents([texts]) + return self.summarize_chain({"input_documents": texts}, return_only_outputs=True)["output_text"] + + def fetch_url_content(self, url): + response = requests.get(url) + soup = BeautifulSoup(response.text, 'html.parser') + + # 提取所有的文本 + text = ''.join(s.getText() for s in soup.find_all('p')) + logging.info(f"Extracted text from {url}") + return text + + def summary_url(self, url): + text = self.fetch_url_content(url) + if text == "": + return "URL unavailable." + text_summary = self.summary(text) + url_content = "webpage content summary:\n" + text_summary + + return url_content + + def ask_url(self, url, question): + text = self.fetch_url_content(url) + if text == "": + return "URL unavailable." + texts = Document(page_content=text) + texts = self.text_splitter.split_documents([texts]) + # use embedding + embeddings = OpenAIEmbeddings(openai_api_key=self.api_key, openai_api_base=os.environ.get("OPENAI_API_BASE", None)) + + # create vectorstore + db = FAISS.from_documents(texts, embeddings) + retriever = db.as_retriever() + qa = RetrievalQA.from_chain_type(llm=self.cheap_llm, chain_type="stuff", retriever=retriever) + return qa.run(f"{question} Reply in 中文") + + def get_answer_at_once(self): + question = self.history[-1]["content"] + # llm=ChatOpenAI(temperature=0, model_name="gpt-3.5-turbo") + agent = initialize_agent(self.tools, self.llm, agent=AgentType.STRUCTURED_CHAT_ZERO_SHOT_REACT_DESCRIPTION, verbose=True) + reply = agent.run(input=f"{question} Reply in 简体中文") + return reply, -1 + + def get_answer_stream_iter(self): + question = self.history[-1]["content"] + it = CallbackToIterator() + manager = BaseCallbackManager(handlers=[ChuanhuCallbackHandler(it.callback)]) + def thread_func(): + tools = self.tools + if self.index is not None: + tools.append( + Tool.from_function( + func=self.query_index, + name="Query Knowledge Base", + description=f"useful when you need to know about: {self.index_summary}", + args_schema=WebBrowsingInput + ) + ) + agent = initialize_agent(self.tools, self.llm, agent=AgentType.STRUCTURED_CHAT_ZERO_SHOT_REACT_DESCRIPTION, verbose=True, callback_manager=manager) + try: + reply = agent.run(input=f"{question} Reply in 简体中文") + except Exception as e: + import traceback + traceback.print_exc() + reply = str(e) + it.callback(reply) + it.finish() + t = Thread(target=thread_func) + t.start() + partial_text = "" + for value in it: + partial_text += value + yield partial_text diff --git a/modules/models/Claude.py b/modules/models/Claude.py new file mode 100644 index 0000000000000000000000000000000000000000..719d1af3a0443ab8510971845c62ce961a13933b --- /dev/null +++ b/modules/models/Claude.py @@ -0,0 +1,55 @@ + +from anthropic import Anthropic, HUMAN_PROMPT, AI_PROMPT +from ..presets import * +from ..utils import * + +from .base_model import BaseLLMModel + + +class Claude_Client(BaseLLMModel): + def __init__(self, model_name, api_secret) -> None: + super().__init__(model_name=model_name) + self.api_secret = api_secret + if None in [self.api_secret]: + raise Exception("请在配置文件或者环境变量中设置Claude的API Secret") + self.claude_client = Anthropic(api_key=self.api_secret) + + + def get_answer_stream_iter(self): + system_prompt = self.system_prompt + history = self.history + if system_prompt is not None: + history = [construct_system(system_prompt), *history] + + completion = self.claude_client.completions.create( + model=self.model_name, + max_tokens_to_sample=300, + prompt=f"{HUMAN_PROMPT}{history}{AI_PROMPT}", + stream=True, + ) + if completion is not None: + partial_text = "" + for chunk in completion: + partial_text += chunk.completion + yield partial_text + else: + yield STANDARD_ERROR_MSG + GENERAL_ERROR_MSG + + + def get_answer_at_once(self): + system_prompt = self.system_prompt + history = self.history + if system_prompt is not None: + history = [construct_system(system_prompt), *history] + + completion = self.claude_client.completions.create( + model=self.model_name, + max_tokens_to_sample=300, + prompt=f"{HUMAN_PROMPT}{history}{AI_PROMPT}", + ) + if completion is not None: + return completion.completion, len(completion.completion) + else: + return "获取资源错误", 0 + + diff --git a/modules/models/DALLE3.py b/modules/models/DALLE3.py new file mode 100644 index 0000000000000000000000000000000000000000..1adca4c990c78f3747c629d0685282d0f6322a48 --- /dev/null +++ b/modules/models/DALLE3.py @@ -0,0 +1,38 @@ +import re +import json +import openai +from openai import OpenAI +from .base_model import BaseLLMModel +from .. import shared +from ..config import retrieve_proxy + + +class OpenAI_DALLE3_Client(BaseLLMModel): + def __init__(self, model_name, api_key, user_name="") -> None: + super().__init__(model_name=model_name, user=user_name) + self.api_key = api_key + + def _get_dalle3_prompt(self): + prompt = self.history[-1]["content"] + if prompt.endswith("--raw"): + prompt = "I NEED to test how the tool works with extremely simple prompts. DO NOT add any detail, just use it AS-IS:" + prompt + return prompt + + @shared.state.switching_api_key + def get_answer_at_once(self): + prompt = self._get_dalle3_prompt() + with retrieve_proxy(): + client = OpenAI(api_key=openai.api_key) + try: + response = client.images.generate( + model="dall-e-3", + prompt=prompt, + size="1024x1024", + quality="standard", + n=1, + ) + except openai.BadRequestError as e: + msg = str(e) + match = re.search(r"'message': '([^']*)'", msg) + return match.group(1), 0 + return f'
{action_name}: {action_input}\n
' + else: + return "" + + +class ChuanhuCallbackHandler(BaseCallbackHandler): + def __init__(self, callback) -> None: + """Initialize callback handler.""" + self.callback = callback + + def on_agent_action( + self, action: AgentAction, color: Optional[str] = None, **kwargs: Any + ) -> Any: + self.callback(get_action_description(action.log)) + + def on_tool_end( + self, + output: str, + color: Optional[str] = None, + observation_prefix: Optional[str] = None, + llm_prefix: Optional[str] = None, + **kwargs: Any, + ) -> None: + """If not the final action, print out observation.""" + # if observation_prefix is not None: + # self.callback(f"\n\n{observation_prefix}") + # self.callback(output) + # if llm_prefix is not None: + # self.callback(f"\n\n{llm_prefix}") + if observation_prefix is not None: + logging.info(observation_prefix) + self.callback(output) + if llm_prefix is not None: + logging.info(llm_prefix) + + def on_agent_finish( + self, finish: AgentFinish, color: Optional[str] = None, **kwargs: Any + ) -> None: + # self.callback(f"{finish.log}\n\n") + logging.info(finish.log) + + def on_llm_new_token(self, token: str, **kwargs: Any) -> None: + """Run on new LLM token. Only available when streaming is enabled.""" + self.callback(token) + + def on_chat_model_start( + self, + serialized: Dict[str, Any], + messages: List[List[BaseMessage]], + **kwargs: Any, + ) -> Any: + """Run when a chat model starts running.""" + pass + + +class ModelType(Enum): + Unknown = -1 + OpenAI = 0 + ChatGLM = 1 + LLaMA = 2 + XMChat = 3 + StableLM = 4 + MOSS = 5 + YuanAI = 6 + Minimax = 7 + ChuanhuAgent = 8 + GooglePaLM = 9 + LangchainChat = 10 + Midjourney = 11 + Spark = 12 + OpenAIInstruct = 13 + Claude = 14 + Qwen = 15 + OpenAIVision = 16 + ERNIE = 17 + DALLE3 = 18 + + @classmethod + def get_type(cls, model_name: str): + model_type = None + model_name_lower = model_name.lower() + if "gpt" in model_name_lower: + if "instruct" in model_name_lower: + model_type = ModelType.OpenAIInstruct + elif "vision" in model_name_lower: + model_type = ModelType.OpenAIVision + else: + model_type = ModelType.OpenAI + elif "chatglm" in model_name_lower: + model_type = ModelType.ChatGLM + elif "llama" in model_name_lower or "alpaca" in model_name_lower: + model_type = ModelType.LLaMA + elif "xmchat" in model_name_lower: + model_type = ModelType.XMChat + elif "stablelm" in model_name_lower: + model_type = ModelType.StableLM + elif "moss" in model_name_lower: + model_type = ModelType.MOSS + elif "yuanai" in model_name_lower: + model_type = ModelType.YuanAI + elif "minimax" in model_name_lower: + model_type = ModelType.Minimax + elif "川虎助理" in model_name_lower: + model_type = ModelType.ChuanhuAgent + elif "palm" in model_name_lower: + model_type = ModelType.GooglePaLM + elif "midjourney" in model_name_lower: + model_type = ModelType.Midjourney + elif "azure" in model_name_lower or "api" in model_name_lower: + model_type = ModelType.LangchainChat + elif "星火大模型" in model_name_lower: + model_type = ModelType.Spark + elif "claude" in model_name_lower: + model_type = ModelType.Claude + elif "qwen" in model_name_lower: + model_type = ModelType.Qwen + elif "ernie" in model_name_lower: + model_type = ModelType.ERNIE + elif "dall" in model_name_lower: + model_type = ModelType.DALLE3 + else: + model_type = ModelType.LLaMA + return model_type + + +class BaseLLMModel: + def __init__( + self, + model_name, + system_prompt=INITIAL_SYSTEM_PROMPT, + temperature=1.0, + top_p=1.0, + n_choices=1, + stop="", + max_generation_token=None, + presence_penalty=0, + frequency_penalty=0, + logit_bias=None, + user="", + single_turn=False, + ) -> None: + self.history = [] + self.all_token_counts = [] + if model_name in MODEL_METADATA: + self.model_name = MODEL_METADATA[model_name]["model_name"] + else: + self.model_name = model_name + self.model_type = ModelType.get_type(model_name) + try: + self.token_upper_limit = MODEL_METADATA[model_name]["token_limit"] + except KeyError: + self.token_upper_limit = DEFAULT_TOKEN_LIMIT + self.interrupted = False + self.system_prompt = system_prompt + self.api_key = None + self.need_api_key = False + self.history_file_path = get_first_history_name(user) + self.user_name = user + self.chatbot = [] + + self.default_single_turn = single_turn + self.default_temperature = temperature + self.default_top_p = top_p + self.default_n_choices = n_choices + self.default_stop_sequence = stop + self.default_max_generation_token = max_generation_token + self.default_presence_penalty = presence_penalty + self.default_frequency_penalty = frequency_penalty + self.default_logit_bias = logit_bias + self.default_user_identifier = user + + self.single_turn = single_turn + self.temperature = temperature + self.top_p = top_p + self.n_choices = n_choices + self.stop_sequence = stop + self.max_generation_token = max_generation_token + self.presence_penalty = presence_penalty + self.frequency_penalty = frequency_penalty + self.logit_bias = logit_bias + self.user_identifier = user + + self.metadata = {} + + def get_answer_stream_iter(self): + """Implement stream prediction. + Conversations are stored in self.history, with the most recent question in OpenAI format. + Should return a generator that yields the next word (str) in the answer. + """ + logging.warning( + "Stream prediction is not implemented. Using at once prediction instead." + ) + response, _ = self.get_answer_at_once() + yield response + + def get_answer_at_once(self): + """predict at once, need to be implemented + conversations are stored in self.history, with the most recent question, in OpenAI format + Should return: + the answer (str) + total token count (int) + """ + logging.warning("at once predict not implemented, using stream predict instead") + response_iter = self.get_answer_stream_iter() + count = 0 + for response in response_iter: + count += 1 + return response, sum(self.all_token_counts) + count + + def billing_info(self): + """get billing infomation, inplement if needed""" + # logging.warning("billing info not implemented, using default") + return BILLING_NOT_APPLICABLE_MSG + + def count_token(self, user_input): + """get token count from input, implement if needed""" + # logging.warning("token count not implemented, using default") + return len(user_input) + + def stream_next_chatbot(self, inputs, chatbot, fake_input=None, display_append=""): + def get_return_value(): + return chatbot, status_text + + status_text = i18n("开始实时传输回答……") + if fake_input: + chatbot.append((fake_input, "")) + else: + chatbot.append((inputs, "")) + + user_token_count = self.count_token(inputs) + self.all_token_counts.append(user_token_count) + logging.debug(f"输入token计数: {user_token_count}") + + stream_iter = self.get_answer_stream_iter() + + if display_append: + display_append = ( + '\n\n' + display_append + ) + partial_text = "" + token_increment = 1 + for partial_text in stream_iter: + if type(partial_text) == tuple: + partial_text, token_increment = partial_text + chatbot[-1] = (chatbot[-1][0], partial_text + display_append) + self.all_token_counts[-1] += token_increment + status_text = self.token_message() + yield get_return_value() + if self.interrupted: + self.recover() + break + self.history.append(construct_assistant(partial_text)) + + def next_chatbot_at_once(self, inputs, chatbot, fake_input=None, display_append=""): + if fake_input: + chatbot.append((fake_input, "")) + else: + chatbot.append((inputs, "")) + if fake_input is not None: + user_token_count = self.count_token(fake_input) + else: + user_token_count = self.count_token(inputs) + self.all_token_counts.append(user_token_count) + ai_reply, total_token_count = self.get_answer_at_once() + self.history.append(construct_assistant(ai_reply)) + if fake_input is not None: + self.history[-2] = construct_user(fake_input) + chatbot[-1] = (chatbot[-1][0], ai_reply + display_append) + if fake_input is not None: + self.all_token_counts[-1] += count_token(construct_assistant(ai_reply)) + else: + self.all_token_counts[-1] = total_token_count - sum(self.all_token_counts) + status_text = self.token_message() + return chatbot, status_text + + def handle_file_upload(self, files, chatbot, language): + """if the model accepts multi modal input, implement this function""" + status = gr.Markdown.update() + if files: + index = construct_index(self.api_key, file_src=files) + status = i18n("索引构建完成") + return gr.Files.update(), chatbot, status + + def summarize_index(self, files, chatbot, language): + status = gr.Markdown.update() + if files: + index = construct_index(self.api_key, file_src=files) + status = i18n("总结完成") + logging.info(i18n("生成内容总结中……")) + os.environ["OPENAI_API_KEY"] = self.api_key + from langchain.chains.summarize import load_summarize_chain + from langchain.prompts import PromptTemplate + from langchain.chat_models import ChatOpenAI + from langchain.callbacks import StdOutCallbackHandler + + prompt_template = ( + "Write a concise summary of the following:\n\n{text}\n\nCONCISE SUMMARY IN " + + language + + ":" + ) + PROMPT = PromptTemplate(template=prompt_template, input_variables=["text"]) + llm = ChatOpenAI() + chain = load_summarize_chain( + llm, + chain_type="map_reduce", + return_intermediate_steps=True, + map_prompt=PROMPT, + combine_prompt=PROMPT, + ) + summary = chain( + {"input_documents": list(index.docstore.__dict__["_dict"].values())}, + return_only_outputs=True, + )["output_text"] + print(i18n("总结") + f": {summary}") + chatbot.append([i18n("上传了") + str(len(files)) + "个文件", summary]) + return chatbot, status + + def prepare_inputs( + self, + real_inputs, + use_websearch, + files, + reply_language, + chatbot, + load_from_cache_if_possible=True, + ): + display_append = [] + limited_context = False + if type(real_inputs) == list: + fake_inputs = real_inputs[0]["text"] + else: + fake_inputs = real_inputs + if files: + from langchain.embeddings.huggingface import HuggingFaceEmbeddings + from langchain.vectorstores.base import VectorStoreRetriever + + limited_context = True + msg = "加载索引中……" + logging.info(msg) + index = construct_index( + self.api_key, + file_src=files, + load_from_cache_if_possible=load_from_cache_if_possible, + ) + assert index is not None, "获取索引失败" + msg = "索引获取成功，生成回答中……" + logging.info(msg) + with retrieve_proxy(): + retriever = VectorStoreRetriever( + vectorstore=index, search_type="similarity", search_kwargs={"k": 6} + ) + # retriever = VectorStoreRetriever(vectorstore=index, search_type="similarity_score_threshold", search_kwargs={ + # "k": 6, "score_threshold": 0.2}) + try: + relevant_documents = retriever.get_relevant_documents(fake_inputs) + except AssertionError: + return self.prepare_inputs( + fake_inputs, + use_websearch, + files, + reply_language, + chatbot, + load_from_cache_if_possible=False, + ) + reference_results = [ + [d.page_content.strip("�"), os.path.basename(d.metadata["source"])] + for d in relevant_documents + ] + reference_results = add_source_numbers(reference_results) + display_append = add_details(reference_results) + display_append = "\n\n" + "".join(display_append) + if type(real_inputs) == list: + real_inputs[0]["text"] = ( + replace_today(PROMPT_TEMPLATE) + .replace("{query_str}", fake_inputs) + .replace("{context_str}", "\n\n".join(reference_results)) + .replace("{reply_language}", reply_language) + ) + else: + real_inputs = ( + replace_today(PROMPT_TEMPLATE) + .replace("{query_str}", real_inputs) + .replace("{context_str}", "\n\n".join(reference_results)) + .replace("{reply_language}", reply_language) + ) + elif use_websearch: + search_results = [] + with DDGS() as ddgs: + ddgs_gen = ddgs.text(fake_inputs, backend="lite") + for r in islice(ddgs_gen, 10): + search_results.append(r) + reference_results = [] + for idx, result in enumerate(search_results): + logging.debug(f"搜索结果{idx + 1}：{result}") + domain_name = urllib3.util.parse_url(result["href"]).host + reference_results.append([result["body"], result["href"]]) + display_append.append( + # f"{idx+1}. [{domain_name}]({result['href']})\n" + f"{idx+1}. {result['title']}" + ) + reference_results = add_source_numbers(reference_results) + # display_append = "
- \n\n" + "".join(display_append) + "
' + "".join(display_append) + "
"
+ )
+ if type(real_inputs) == list:
+ real_inputs[0]["text"] = (
+ replace_today(WEBSEARCH_PTOMPT_TEMPLATE)
+ .replace("{query}", fake_inputs)
+ .replace("{web_results}", "\n\n".join(reference_results))
+ .replace("{reply_language}", reply_language)
+ )
+ else:
+ real_inputs = (
+ replace_today(WEBSEARCH_PTOMPT_TEMPLATE)
+ .replace("{query}", fake_inputs)
+ .replace("{web_results}", "\n\n".join(reference_results))
+ .replace("{reply_language}", reply_language)
+ )
+ else:
+ display_append = ""
+ return limited_context, fake_inputs, display_append, real_inputs, chatbot
+
+ def predict(
+ self,
+ inputs,
+ chatbot,
+ stream=False,
+ use_websearch=False,
+ files=None,
+ reply_language="中文",
+ should_check_token_count=True,
+ ): # repetition_penalty, top_k
+ status_text = "开始生成回答……"
+ if type(inputs) == list:
+ logging.info(
+ "用户"
+ + f"{self.user_name}"
+ + "的输入为："
+ + colorama.Fore.BLUE
+ + "("
+ + str(len(inputs) - 1)
+ + " images) "
+ + f"{inputs[0]['text']}"
+ + colorama.Style.RESET_ALL
+ )
+ else:
+ logging.info(
+ "用户"
+ + f"{self.user_name}"
+ + "的输入为："
+ + colorama.Fore.BLUE
+ + f"{inputs}"
+ + colorama.Style.RESET_ALL
+ )
+ if should_check_token_count:
+ if type(inputs) == list:
+ yield chatbot + [(inputs[0]["text"], "")], status_text
+ else:
+ yield chatbot + [(inputs, "")], status_text
+ if reply_language == "跟随问题语言（不稳定）":
+ reply_language = "the same language as the question, such as English, 中文, 日本語, Español, Français, or Deutsch."
+
+ (
+ limited_context,
+ fake_inputs,
+ display_append,
+ inputs,
+ chatbot,
+ ) = self.prepare_inputs(
+ real_inputs=inputs,
+ use_websearch=use_websearch,
+ files=files,
+ reply_language=reply_language,
+ chatbot=chatbot,
+ )
+ yield chatbot + [(fake_inputs, "")], status_text
+
+ if (
+ self.need_api_key
+ and self.api_key is None
+ and not shared.state.multi_api_key
+ ):
+ status_text = STANDARD_ERROR_MSG + NO_APIKEY_MSG
+ logging.info(status_text)
+ chatbot.append((fake_inputs, ""))
+ if len(self.history) == 0:
+ self.history.append(construct_user(fake_inputs))
+ self.history.append("")
+ self.all_token_counts.append(0)
+ else:
+ self.history[-2] = construct_user(fake_inputs)
+ yield chatbot + [(fake_inputs, "")], status_text
+ return
+ elif len(fake_inputs.strip()) == 0:
+ status_text = STANDARD_ERROR_MSG + NO_INPUT_MSG
+ logging.info(status_text)
+ yield chatbot + [(fake_inputs, "")], status_text
+ return
+
+ if self.single_turn:
+ self.history = []
+ self.all_token_counts = []
+ if type(inputs) == list:
+ self.history.append(inputs)
+ else:
+ self.history.append(construct_user(inputs))
+
+ try:
+ if stream:
+ logging.debug("使用流式传输")
+ iter = self.stream_next_chatbot(
+ inputs,
+ chatbot,
+ fake_input=fake_inputs,
+ display_append=display_append,
+ )
+ for chatbot, status_text in iter:
+ yield chatbot, status_text
+ else:
+ logging.debug("不使用流式传输")
+ chatbot, status_text = self.next_chatbot_at_once(
+ inputs,
+ chatbot,
+ fake_input=fake_inputs,
+ display_append=display_append,
+ )
+ yield chatbot, status_text
+ except Exception as e:
+ traceback.print_exc()
+ status_text = STANDARD_ERROR_MSG + beautify_err_msg(str(e))
+ yield chatbot, status_text
+
+ if len(self.history) > 1 and self.history[-1]["content"] != fake_inputs:
+ logging.info(
+ "回答为："
+ + colorama.Fore.BLUE
+ + f"{self.history[-1]['content']}"
+ + colorama.Style.RESET_ALL
+ )
+
+ if limited_context:
+ # self.history = self.history[-4:]
+ # self.all_token_counts = self.all_token_counts[-2:]
+ self.history = []
+ self.all_token_counts = []
+
+ max_token = self.token_upper_limit - TOKEN_OFFSET
+
+ if sum(self.all_token_counts) > max_token and should_check_token_count:
+ count = 0
+ while (
+ sum(self.all_token_counts)
+ > self.token_upper_limit * REDUCE_TOKEN_FACTOR
+ and sum(self.all_token_counts) > 0
+ ):
+ count += 1
+ del self.all_token_counts[0]
+ del self.history[:2]
+ logging.info(status_text)
+ status_text = f"为了防止token超限，模型忘记了早期的 {count} 轮对话"
+ yield chatbot, status_text
+
+ self.chatbot = chatbot
+ self.auto_save(chatbot)
+
+ def retry(
+ self,
+ chatbot,
+ stream=False,
+ use_websearch=False,
+ files=None,
+ reply_language="中文",
+ ):
+ logging.debug("重试中……")
+ if len(self.history) > 1:
+ inputs = self.history[-2]["content"]
+ del self.history[-2:]
+ if len(self.all_token_counts) > 0:
+ self.all_token_counts.pop()
+ elif len(chatbot) > 0:
+ inputs = chatbot[-1][0]
+ if '