writinwaters
commited on
Commit
·
99b8bda
1
Parent(s):
b0b89da
0328 restructure readme (#165)
Browse files* Added emojis
* Miscellaneous editorial updates
README.md
CHANGED
@@ -20,70 +20,66 @@
|
|
20 |
<img height="21" src="https://img.shields.io/badge/License-Apache--2.0-ffffff?style=flat-square&labelColor=d4eaf7&color=7d09f1" alt="license">
|
21 |
</a>
|
22 |
</p>
|
23 |
-
|
24 |
-
[RagFlow](http://demo.ragflow.io) is a knowledge management platform built on custom-build document understanding engine and LLM,
|
25 |
-
with reasoned and well-founded answers to your question. Clone this repository, you can deploy your own knowledge management
|
26 |
-
platform to empower your business with AI.
|
27 |
|
|
|
28 |
<div align="center" style="margin-top:20px;margin-bottom:20px;">
|
29 |
<img src="https://github.com/infiniflow/ragflow/assets/12318111/b24a7a5f-4d1d-4a30-90b1-7b0ec558b79d" width="1000"/>
|
30 |
</div>
|
31 |
|
32 |
-
|
33 |
-
-
|
34 |
- For documents from different domain for different purpose, the engine applys different analyzing and search strategy.
|
35 |
- Easily intervene and manipulate the data proccessing procedure when things goes beyond expectation.
|
36 |
- Multi-media document understanding is supported using OCR and multi-modal LLM.
|
37 |
-
-
|
38 |
- For PDF files, layout and table structures including row, column and span of them are recognized.
|
39 |
- Put the table accrossing the pages together.
|
40 |
- Reconstruct the table structure components into html table.
|
41 |
- **Querying database dumped data are supported.** After uploading tables from any database, you can search any data records just by asking.
|
42 |
-
-
|
43 |
-
- The record number uploaded is not limited.
|
44 |
-
- Some extra description of column headers should be provided.
|
45 |
- **Reasoned and well-founded answers.** The cited document part in LLM's answer is provided and pointed out in the original document.
|
46 |
-
- The answers are based on retrieved result for which we apply vector-keyword hybrids search and
|
47 |
- The part of document cited in the answer is presented in the most expressive way.
|
48 |
- For PDF file, the cited parts in document can be located in the original PDF.
|
49 |
-
|
50 |
|
51 |
-
|
52 |
-
|
53 |
-
|
|
|
|
|
54 |
|
55 |
-
|
56 |
-
## System Requirements
|
57 |
-
Be aware of the system minimum requirements before starting installation.
|
58 |
- CPU >= 2 cores
|
59 |
-
- RAM >=
|
60 |
|
61 |
Then, you need to check the following command:
|
62 |
```bash
|
63 |
121:/ragflow# sysctl vm.max_map_count
|
64 |
vm.max_map_count = 262144
|
65 |
```
|
66 |
-
If **vm.max_map_count** is not
|
67 |
```bash
|
68 |
121:/ragflow# sudo sysctl -w vm.max_map_count=262144
|
69 |
```
|
70 |
-
|
71 |
-
|
72 |
-
Add or update the following line in the file:
|
73 |
```bash
|
74 |
vm.max_map_count=262144
|
75 |
```
|
76 |
|
77 |
-
|
78 |
|
79 |
-
If
|
80 |
|
81 |
-
|
82 |
|
83 |
> - In [service_conf.yaml](./docker/service_conf.yaml), configuration of *LLM* in **user_default_llm** is strongly recommended.
|
84 |
> In **user_default_llm** of [service_conf.yaml](./docker/service_conf.yaml), you need to specify LLM factory and your own _API_KEY_.
|
85 |
-
>
|
86 |
-
|
|
|
87 |
> [OpenAI](https://platform.openai.com/login?launch), [Tongyi-Qianwen](https://dashscope.console.aliyun.com/model),
|
88 |
> [ZHIPU-AI](https://open.bigmodel.cn/), [Moonshot](https://platform.moonshot.cn/docs/docs)
|
89 |
```bash
|
@@ -100,13 +96,15 @@ $ docker build -t infiniflow/ragflow:v1.0 .
|
|
100 |
$ cd ragflow/docker
|
101 |
$ docker compose up -d
|
102 |
```
|
103 |
-
> The core image is about
|
104 |
|
105 |
-
|
106 |
-
_**Hallelujah!**_ You have successfully launched the system.
|
107 |
```bash
|
108 |
$ docker logs -f ragflow-server
|
|
|
|
|
109 |
|
|
|
110 |
____ ______ __
|
111 |
/ __ \ ____ _ ____ _ / ____// /____ _ __
|
112 |
/ /_/ // __ `// __ `// /_ / // __ \| | /| / /
|
@@ -121,35 +119,36 @@ INFO:werkzeug:Press CTRL+C to quit
|
|
121 |
|
122 |
```
|
123 |
Open your browser, enter the IP address of your server, _**Hallelujah**_ again!
|
124 |
-
> The default serving port is 80, if you want to change that,
|
125 |
-
> and change the left part of *'80:80'*'.
|
126 |
|
127 |
-
|
128 |
|
129 |
<div align="center" style="margin-top:20px;margin-bottom:20px;">
|
130 |
<img src="https://github.com/infiniflow/ragflow/assets/12318111/d6ac5664-c237-4200-a7c2-a4a00691b485" width="1000"/>
|
131 |
</div>
|
|
|
132 |
|
133 |
-
# Configuration
|
134 |
If you need to change the default setting of the system when you deploy it. There several ways to configure it.
|
135 |
-
Please refer to [README](./docker/README.md)
|
136 |
After changing something, please run *docker-compose up -d* again.
|
137 |
|
138 |
-
> If you want to change the basic setups, like port, password .etc., please refer to [.env](./docker/.env) before starting the system.
|
139 |
|
140 |
-
> If you change anything in [.env](./docker/.env), please check [service_conf.yaml](./docker/service_conf.yaml) which is a
|
141 |
-
> configuration of the back-end service and should be consistent with [.env](./docker/.env).
|
142 |
|
143 |
-
|
144 |
|
145 |
-
|
146 |
-
- [ ] Support URLs. Crawl web and extract the main content.
|
147 |
|
|
|
148 |
|
149 |
-
|
|
|
|
|
|
|
|
|
150 |
|
151 |
-
For those who'd like to contribute code, see our [Contribution Guide](https://github.com/infiniflow/ragflow/blob/main/CONTRIBUTING.md).
|
152 |
|
153 |
-
|
154 |
|
155 |
-
|
|
|
20 |
<img height="21" src="https://img.shields.io/badge/License-Apache--2.0-ffffff?style=flat-square&labelColor=d4eaf7&color=7d09f1" alt="license">
|
21 |
</a>
|
22 |
</p>
|
23 |
+
[RagFlow](http://demo.ragflow.io) is a knowledge management platform built on custom-build document understanding engine and LLM, with reasoned and well-founded answers to your question. Clone this repository, you can deploy your own knowledge management platform to empower your business with AI.
|
|
|
|
|
|
|
24 |
|
25 |
+
|
26 |
<div align="center" style="margin-top:20px;margin-bottom:20px;">
|
27 |
<img src="https://github.com/infiniflow/ragflow/assets/12318111/b24a7a5f-4d1d-4a30-90b1-7b0ec558b79d" width="1000"/>
|
28 |
</div>
|
29 |
|
30 |
+
## 🌟Key Features
|
31 |
+
- 🍭**Custom-build document understanding engine.** Our deep learning engine is made according to the needs of analyzing and searching various type of documents in different domain.
|
32 |
- For documents from different domain for different purpose, the engine applys different analyzing and search strategy.
|
33 |
- Easily intervene and manipulate the data proccessing procedure when things goes beyond expectation.
|
34 |
- Multi-media document understanding is supported using OCR and multi-modal LLM.
|
35 |
+
- 🍭**State-of-the-art table structure and layout recognition.** Precisely extract and understand the document including table content. See [README.](./deepdoc/README.md)
|
36 |
- For PDF files, layout and table structures including row, column and span of them are recognized.
|
37 |
- Put the table accrossing the pages together.
|
38 |
- Reconstruct the table structure components into html table.
|
39 |
- **Querying database dumped data are supported.** After uploading tables from any database, you can search any data records just by asking.
|
40 |
+
- You can now query a database using natural language instead of using SQL.
|
41 |
+
- The record number uploaded is not limited.
|
|
|
42 |
- **Reasoned and well-founded answers.** The cited document part in LLM's answer is provided and pointed out in the original document.
|
43 |
+
- The answers are based on retrieved result for which we apply vector-keyword hybrids search and re-rank.
|
44 |
- The part of document cited in the answer is presented in the most expressive way.
|
45 |
- For PDF file, the cited parts in document can be located in the original PDF.
|
|
|
46 |
|
47 |
+
## 🤺RagFlow vs. other RAG applications
|
48 |
+
|
49 |
+
|
50 |
+
|
51 |
+
## 🎬 Get Started
|
52 |
|
53 |
+
### 📝Prerequisites
|
|
|
|
|
54 |
- CPU >= 2 cores
|
55 |
+
- RAM >= 8 GB
|
56 |
|
57 |
Then, you need to check the following command:
|
58 |
```bash
|
59 |
121:/ragflow# sysctl vm.max_map_count
|
60 |
vm.max_map_count = 262144
|
61 |
```
|
62 |
+
If **vm.max_map_count** is not greater than 65535:
|
63 |
```bash
|
64 |
121:/ragflow# sudo sysctl -w vm.max_map_count=262144
|
65 |
```
|
66 |
+
Note that this change is reset after a system reboot. To render your change permanent, add or update the following line in **/etc/sysctl.conf**:
|
67 |
+
|
|
|
68 |
```bash
|
69 |
vm.max_map_count=262144
|
70 |
```
|
71 |
|
72 |
+
### Install docker
|
73 |
|
74 |
+
If you have not installed *Docker* on your local machine, see [Install Docker Engine](https://docs.docker.com/engine/install/)
|
75 |
|
76 |
+
### Quick Start
|
77 |
|
78 |
> - In [service_conf.yaml](./docker/service_conf.yaml), configuration of *LLM* in **user_default_llm** is strongly recommended.
|
79 |
> In **user_default_llm** of [service_conf.yaml](./docker/service_conf.yaml), you need to specify LLM factory and your own _API_KEY_.
|
80 |
+
> If you do not have _API_KEY_ at the moment, you can specify it in
|
81 |
+
Settings the next time you log in to the system.
|
82 |
+
> - RagFlow supports the flowing LLM factory, with more coming in the pipeline:
|
83 |
> [OpenAI](https://platform.openai.com/login?launch), [Tongyi-Qianwen](https://dashscope.console.aliyun.com/model),
|
84 |
> [ZHIPU-AI](https://open.bigmodel.cn/), [Moonshot](https://platform.moonshot.cn/docs/docs)
|
85 |
```bash
|
|
|
96 |
$ cd ragflow/docker
|
97 |
$ docker compose up -d
|
98 |
```
|
99 |
+
> The core image is about 15 GB in size and may take a while to load.
|
100 |
|
101 |
+
Check the server status after pulling all images and running up:
|
|
|
102 |
```bash
|
103 |
$ docker logs -f ragflow-server
|
104 |
+
```
|
105 |
+
*Hallelujah! The following outputs indicates that you have successfully launched the system:*
|
106 |
|
107 |
+
```bash
|
108 |
____ ______ __
|
109 |
/ __ \ ____ _ ____ _ / ____// /____ _ __
|
110 |
/ /_/ // __ `// __ `// /_ / // __ \| | /| / /
|
|
|
119 |
|
120 |
```
|
121 |
Open your browser, enter the IP address of your server, _**Hallelujah**_ again!
|
122 |
+
> The default serving port is 80, if you want to change that, refer to the [docker-compose.yml](./docker-compose.yaml) and change the left part of *'80:80'*'.
|
|
|
123 |
|
124 |
+
## 🔎System Architecture
|
125 |
|
126 |
<div align="center" style="margin-top:20px;margin-bottom:20px;">
|
127 |
<img src="https://github.com/infiniflow/ragflow/assets/12318111/d6ac5664-c237-4200-a7c2-a4a00691b485" width="1000"/>
|
128 |
</div>
|
129 |
+
## 🔧 Configurations
|
130 |
|
|
|
131 |
If you need to change the default setting of the system when you deploy it. There several ways to configure it.
|
132 |
+
Please refer to this [README](./docker/README.md) to manually update the configuration.
|
133 |
After changing something, please run *docker-compose up -d* again.
|
134 |
|
135 |
+
> If you want to change the basic setups, like port, password .etc., please refer to [.env](./docker/.env) before starting up the system.
|
136 |
|
137 |
+
> If you change anything in [.env](./docker/.env), please check [service_conf.yaml](./docker/service_conf.yaml) which is a configuration of the back-end service and should be consistent with [.env](./docker/.env).
|
|
|
138 |
|
139 |
+
## 📜 Roadmap
|
140 |
|
141 |
+
See the [RagFlow Roadmap 2024](https://github.com/infiniflow/ragflow/issues/162)
|
|
|
142 |
|
143 |
+
## 🏄Community
|
144 |
|
145 |
+
- [Discord](https://discord.gg/uqQ4YMDf)
|
146 |
+
- X
|
147 |
+
- [GitHub Discussions]()
|
148 |
+
- YouTube
|
149 |
+
- WeChat
|
150 |
|
|
|
151 |
|
152 |
+
## 🙌 Contributing
|
153 |
|
154 |
+
For those who'd like to contribute code, see our [Contribution Guide](https://github.com/infiniflow/ragflow/blob/main/CONTRIBUTING.md).
|