writinwaters
commited on
Commit
·
d3d8d70
1
Parent(s):
3313a7c
0329 readme (#175)
Browse files* Updated Get Started.
* Finishing off Get Started
* minor
* minor
README.md
CHANGED
|
@@ -45,22 +45,6 @@
|
|
| 45 |
- The part of document cited in the answer is presented in the most expressive way.
|
| 46 |
- For PDF file, the cited parts in document can be located in the original PDF.
|
| 47 |
|
| 48 |
-
## 🤺RagFlow vs. other RAG applications
|
| 49 |
-
|
| 50 |
-
| Feature | RagFlow | Langchain-Chatchat | Assistants API | QAnythig | LangChain |
|
| 51 |
-
|---------|:---------:|:-----------:|:-----------:|:-----------:|:-----------:|
|
| 52 |
-
| **Well-Founded Answer** | :white_check_mark: | :x: | :x: | :x: | :x: |
|
| 53 |
-
| **Trackable Chunking** | :white_check_mark: | :x: | :x: | :x: | :x: |
|
| 54 |
-
| **Chunking Method** | Rich Variety | Naive | Naive | Naive | Naive |
|
| 55 |
-
| **Table Structure Recognition** | :white_check_mark: | :x: | :x: | :x: | :x: |
|
| 56 |
-
| **Structured Data Lookup** | :white_check_mark: | :x: | :x: | :x: | :x: | :x: |
|
| 57 |
-
| **Programming Approach** | API-oriented | API-oriented | API-oriented | API-oriented | Python Code-oriented |
|
| 58 |
-
| **RAG Engine** | :white_check_mark: | :white_check_mark: | :white_check_mark: | :x: | :x: |
|
| 59 |
-
| **Prompt IDE** | :white_check_mark: | :white_check_mark: | :white_check_mark: | :x: | :x: |
|
| 60 |
-
| **Supported LLMs** | Rich Variety | Rich Variety | OpenAI-only | QwenLLM | Rich Variety |
|
| 61 |
-
| **Local Deployment** | :white_check_mark: | :white_check_mark: | :x: | :x: | :x: |
|
| 62 |
-
| **Ecosystem Strategy** | Open Source | Open Source | Close Source | Open Source | Open Source |
|
| 63 |
-
|
| 64 |
## 🔎 System Architecture
|
| 65 |
|
| 66 |
<div align="center" style="margin-top:20px;margin-bottom:20px;">
|
|
@@ -82,13 +66,14 @@
|
|
| 82 |
> $ sysctl vm.max_map_count
|
| 83 |
> ```
|
| 84 |
>
|
| 85 |
-
> Reset `vm.max_map_count` to a value greater than 65535 if it is not.
|
| 86 |
>
|
| 87 |
> ```bash
|
|
|
|
| 88 |
> $ sudo sysctl -w vm.max_map_count=262144
|
| 89 |
> ```
|
| 90 |
>
|
| 91 |
-
> This change will be reset after a system reboot. To ensure your change remains permanent, add or update the
|
| 92 |
>
|
| 93 |
> ```bash
|
| 94 |
> vm.max_map_count=262144
|
|
@@ -98,7 +83,7 @@
|
|
| 98 |
|
| 99 |
### Start up the RagFlow server
|
| 100 |
|
| 101 |
-
1. Clone the repo
|
| 102 |
|
| 103 |
```bash
|
| 104 |
$ git clone https://github.com/infiniflow/ragflow.git
|
|
@@ -106,36 +91,33 @@
|
|
| 106 |
|
| 107 |
|
| 108 |
|
| 109 |
-
2.
|
| 110 |
|
| 111 |
-
> -
|
| 112 |
-
>
|
| 113 |
-
> If you do not have _API_KEY_ at the moment, you can specify it in
|
| 114 |
-
Settings the next time you log in to the system.
|
| 115 |
-
> - RagFlow supports the flowing LLM factory, with more coming in the pipeline:
|
| 116 |
-
> [OpenAI](https://platform.openai.com/login?launch), [Tongyi-Qianwen](https://dashscope.console.aliyun.com/model),
|
| 117 |
-
> [ZHIPU-AI](https://open.bigmodel.cn/), [Moonshot](https://platform.moonshot.cn/docs/docs)
|
| 118 |
-
```bash
|
| 119 |
|
| 120 |
-
|
| 121 |
-
$ docker compose up -d
|
| 122 |
-
```
|
| 123 |
-
### OR
|
| 124 |
|
| 125 |
-
```bash
|
| 126 |
-
|
| 127 |
-
$ cd ragflow/
|
| 128 |
-
$ docker
|
| 129 |
-
|
| 130 |
-
|
| 131 |
-
|
| 132 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 133 |
|
| 134 |
-
|
| 135 |
-
|
| 136 |
-
|
| 137 |
-
```
|
| 138 |
-
|
|
|
|
|
|
|
| 139 |
|
| 140 |
```bash
|
| 141 |
____ ______ __
|
|
@@ -149,10 +131,9 @@ $ docker logs -f ragflow-server
|
|
| 149 |
* Running on http://127.0.0.1:9380
|
| 150 |
* Running on http://172.22.0.5:9380
|
| 151 |
INFO:werkzeug:Press CTRL+C to quit
|
| 152 |
-
|
| 153 |
```
|
| 154 |
-
In your browser, enter the IP address of your server.
|
| 155 |
|
|
|
|
| 156 |
|
| 157 |
|
| 158 |
## 🔧 Configurations
|
|
|
|
| 45 |
- The part of document cited in the answer is presented in the most expressive way.
|
| 46 |
- For PDF file, the cited parts in document can be located in the original PDF.
|
| 47 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
## 🔎 System Architecture
|
| 49 |
|
| 50 |
<div align="center" style="margin-top:20px;margin-bottom:20px;">
|
|
|
|
| 66 |
> $ sysctl vm.max_map_count
|
| 67 |
> ```
|
| 68 |
>
|
| 69 |
+
> Reset `vm.max_map_count` to a value greater than 65535 if it is not.
|
| 70 |
>
|
| 71 |
> ```bash
|
| 72 |
+
> # In this case, we set it to 262144:
|
| 73 |
> $ sudo sysctl -w vm.max_map_count=262144
|
| 74 |
> ```
|
| 75 |
>
|
| 76 |
+
> This change will be reset after a system reboot. To ensure your change remains permanent, add or update the `vm.max_map_count` value in **/etc/sysctl.conf** accordingly:
|
| 77 |
>
|
| 78 |
> ```bash
|
| 79 |
> vm.max_map_count=262144
|
|
|
|
| 83 |
|
| 84 |
### Start up the RagFlow server
|
| 85 |
|
| 86 |
+
1. Clone the repo:
|
| 87 |
|
| 88 |
```bash
|
| 89 |
$ git clone https://github.com/infiniflow/ragflow.git
|
|
|
|
| 91 |
|
| 92 |
|
| 93 |
|
| 94 |
+
2. **Recommended**: In **docker/service_conf.yaml**, select the desired LLM factory in `user_default_llm` and update the `API_KEY` field with your own.
|
| 95 |
|
| 96 |
+
> - You can still continue with the default settings, but it is highly recommended that you use your own API key the next time you log into the system.
|
| 97 |
+
> - RagFlow now supports the flowing LLM factories: OpenAI, Tongyi-Qianwen, ZHIPU-AI, and Moonshot.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 98 |
|
| 99 |
+
3. You now presented with two options for building the system: Using the pre-built images or building the images from source:
|
|
|
|
|
|
|
|
|
|
| 100 |
|
| 101 |
+
```bash
|
| 102 |
+
# To use the pre-built images:
|
| 103 |
+
$ cd ragflow/docker
|
| 104 |
+
$ docker compose up -d
|
| 105 |
+
```
|
| 106 |
+
```bash
|
| 107 |
+
# To build the images from source:
|
| 108 |
+
$ cd ragflow/
|
| 109 |
+
$ docker build -t infiniflow/ragflow:v1.0 .
|
| 110 |
+
$ cd ragflow/docker
|
| 111 |
+
$ docker compose up -d
|
| 112 |
+
```
|
| 113 |
|
| 114 |
+
> The core image is about 15 GB in size and may take a while to load.
|
| 115 |
+
|
| 116 |
+
4. Check the server status after pulling all images and having Docker up and running:
|
| 117 |
+
```bash
|
| 118 |
+
$ docker logs -f ragflow-server
|
| 119 |
+
```
|
| 120 |
+
*The following output confirms a successful launch of the system:*
|
| 121 |
|
| 122 |
```bash
|
| 123 |
____ ______ __
|
|
|
|
| 131 |
* Running on http://127.0.0.1:9380
|
| 132 |
* Running on http://172.22.0.5:9380
|
| 133 |
INFO:werkzeug:Press CTRL+C to quit
|
|
|
|
| 134 |
```
|
|
|
|
| 135 |
|
| 136 |
+
5. In your browser, enter the IP address of your server and now you can try it out.
|
| 137 |
|
| 138 |
|
| 139 |
## 🔧 Configurations
|