Merge branch 'master' into dev

README.md

@@ -311,7 +311,7 @@ chatchat-config model --default_llm_model qwen2-instruct
 ```shell
 # 这里应为 3.2 中 "CHATCHAT_ROOT" 变量指向目录
 cd /root/anaconda3/envs/chatchat/lib/python3.11/site-packages/chatchat
-vim model_providers.yaml
+vim configs/model_providers.yaml
 ```
 
 配置介绍请参考 [model-providers/README.md](libs/model-providers/README.md)
@@ -427,6 +427,14 @@ chatchat -a
 > ```
 > </details>
 
+### Docker 部署
+```shell
+docker pull chatimage/chatchat:0.3.0-0623-3
+```
+> [!important]
+> 强烈建议: 使用 docker-compose 部署, 具体参考 [README_docker](docs/install/README_docker.md)
+
+
 ### 旧版本迁移
 
 * 0.3.x 结构改变很大,强烈建议您按照文档重新部署. 以下指南不保证100%兼容和成功. 记得提前备份重要数据!

docker/Dockerfile

@@ -1,19 +1,29 @@
+# Base Image
 FROM python:3.11
-
-RUN apt-get update
-RUN apt-get install -y libgl1-mesa-glx
-
-RUN mkdir /Langchain-Chatchat
-COPY requirements.txt /Langchain-Chatchat
-COPY requirements_api.txt /Langchain-Chatchat
-COPY requirements_webui.txt /Langchain-Chatchat
-
-WORKDIR /Langchain-Chatchat
-RUN pip install --upgrade pip
-RUN pip install -r requirements.txt
-RUN pip install -r requirements_api.txt
-RUN pip install -r requirements_webui.txt
-
-EXPOSE 8501
-EXPOSE 7861
-EXPOSE 20000
+# Labels
+LABEL maintainer=chatchat
+# Environment Variables
+ENV HOME=/usr/local/lib/python3.11/site-packages/chatchat
+# Init Environment
+RUN ln -sf /usr/share/zoneinfo/Asia/Shanghai /etc/localtime && \
+    echo "Asia/Shanghai" > /etc/timezone
+# Install Dependencies
+RUN apt-get update -y && \
+    apt-get install -y --no-install-recommends libgl1 libglib2.0-0 && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/*
+RUN pip install openpyxl networkx faiss-cpu jq unstructured[pdf] \
+    opencv-python rapidocr-onnxruntime PyMuPDF rank_bm25 youtube_search python-docx
+# Install Chatchat
+RUN pip install --index-url https://pypi.python.org/simple/ langchain-chatchat -U
+# Install ModelProvider
+RUN pip install xinference-client
+# Make Custom Settings
+RUN chatchat-config server --default_bind_host=0.0.0.0 && \
+    chatchat-config model --default_llm_model qwen2-instruct
+# Copy Data
+COPY /libs/chatchat-server/chatchat/configs/model_providers.yaml $HOME/configs/model_providers.yaml
+ADD /docker/data.tar.gz $HOME/
+WORKDIR $HOME
+EXPOSE 7861 8501
+ENTRYPOINT ["chatchat", "-a"]

docker/docker-compose.yaml

@@ -0,0 +1,38 @@
+version: '3.9'
+services:
+  xinference:
+    image: xprobe/xinference:v0.12.1
+    restart: always
+    command: xinference-local -H 0.0.0.0
+    # ports: # 不使用 host network 时可打开.
+    #  - "9997:9997"
+    network_mode: "host"
+    # 将本地路径(~/xinference)挂载到容器路径(/root/.xinference)中,
+    # 详情见: https://inference.readthedocs.io/zh-cn/latest/getting_started/using_docker_image.html
+    volumes:
+      - ~/xinference:/root/.xinference
+      # - ~/xinference/cache/huggingface:/root/.cache/huggingface
+      # - ~/xinference/cache/modelscope:/root/.cache/modelscope
+    deploy:
+      resources:
+        reservations:
+          devices:
+            - driver: nvidia
+              count: all
+              capabilities: [gpu]
+    runtime: nvidia
+    # 模型源更改为 ModelScope, 默认为 HuggingFace
+    # environment:
+    #   - XINFERENCE_MODEL_SRC=modelscope
+  chatchat:
+    image: chatimage/chatchat:0.3.0-0623-3
+    restart: always
+    # ports: # 不使用 host network 时可打开.
+    #   - "7861:7861"
+    #   - "8501:8501"
+    network_mode: "host"
+    # 将本地路径(~/chatchat/data)挂载到容器默认数据路径(/usr/local/lib/python3.11/site-packages/chatchat/data)中
+    # 将本地模型接入配置文件(~/chatchat/model_providers.yaml)挂载到容器默认模型接入配置文件路径(/usr/local/lib/python3.11/site-packages/chatchat/configs/)中
+    # volumes:
+    #   - ~/chatchat/data:/usr/local/lib/python3.11/site-packages/chatchat/data
+    #   - ~/chatchat/model_providers.yaml:/usr/local/lib/python3.11/site-packages/chatchat/configs/model_providers.yaml

docs/install/README_docker.md

@@ -0,0 +1,183 @@
+### chatchat 容器化部署指引
+
+> 提示: 此指引为在 Linux 环境下编写完成, 其他环境下暂未测试, 理论上可行.
+> 
+> Langchain-Chatchat docker 镜像已支持多架构, 欢迎大家自行测试.
+
+#### 一. Langchain-Chatchat 体验部署
+
+##### 1. 安装 docker-compose
+寻找适合你环境的 docker-compose 版本, 请参考 [Docker-Compose](https://github.com/docker/compose).
+
+举例: Linux X86 环境 可下载 [docker-compose-linux-x86_64](https://github.com/docker/compose/releases/download/v2.27.3/docker-compose-linux-x86_64) 使用.
+```shell
+cd ~
+wget https://github.com/docker/compose/releases/download/v2.27.3/docker-compose-linux-x86_64
+mv docker-compose-linux-x86_64 /usr/bin/docker-compose
+which docker-compose
+```
+/usr/bin/docker-compose
+```shell
+docker-compose -v
+```
+Docker Compose version v2.27.3
+
+##### 2. 安装 NVIDIA Container Toolkit
+寻找适合你环境的 NVIDIA Container Toolkit 版本, 请参考: [Installing the NVIDIA Container Toolkit](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html).
+
+安装完成后记得按照刚刚文档中`Configuring Docker`章节对 docker 进行初始化. 
+
+##### 3. 创建 xinference 数据缓存路径
+
+这一步强烈建议, 因为可以将 xinference 缓存的模型都保存到本地, 长期使用. 
+```shell
+mkdir -p ~/xinference
+```
+
+##### 4. 下载 chatchat & xinference 启动配置文件(docker-compose.yaml)
+```shell
+cd ~
+wget https://github.com/chatchat-space/Langchain-Chatchat/blob/master/docker/docker-compose.yaml
+```
+
+##### 5. 启动 chatchat & xinference 服务
+```shell
+docker-compose up -d
+```
+出现如下日志即为成功 ( 第一次启动需要下载 docker 镜像, 时间较长, 这里已经提前下载好了 )
+```text
+WARN[0000] /root/docker-compose.yaml: `version` is obsolete 
+[+] Running 2/2
+ ✔ Container root-chatchat-1    Started                                                                                             0.2s 
+ ✔ Container root-xinference-1  Started                                                                                             0.3s
+```
+
+##### 6.检查服务启动情况
+```shell
+docker-compose up -d
+```
+```text
+WARN[0000] /root/docker-compose.yaml: `version` is obsolete 
+NAME                IMAGE                           COMMAND                  SERVICE      CREATED         STATUS         PORTS
+root-chatchat-1     chatimage/chatchat:0.3.0-0622   "chatchat -a"            chatchat     3 minutes ago   Up 3 minutes   
+root-xinference-1   xprobe/xinference:v0.12.1       "/opt/nvidia/nvidia_…"   xinference   3 minutes ago   Up 3 minutes
+```
+```shell
+ss -anptl | grep -E '(8501|7861|9997)'
+```
+```text
+LISTEN 0      128          0.0.0.0:9997       0.0.0.0:*    users:(("pt_main_thread",pid=1489804,fd=21))
+LISTEN 0      128          0.0.0.0:8501       0.0.0.0:*    users:(("python",pid=1490078,fd=10))        
+LISTEN 0      128          0.0.0.0:7861       0.0.0.0:*    users:(("python",pid=1490014,fd=9))
+```
+如上, 服务均已正常启动, 即可体验使用.
+
+> 提示: 先登陆 xinference ui `http://<your_ip>:9997` 启动 llm 和 embedding 后, 再登陆 chatchat ui `http://<your_ip>:8501` 进行体验.
+> 
+> 详细文档:
+> - Langchain-chatchat 使用请参考: [LangChain-Chatchat](/README.md)
+> 
+> - Xinference 使用请参考: [欢迎来到 Xinference！](https://inference.readthedocs.io/zh-cn/latest/index.html)
+
+#### 二. Langchain-Chatchat 进阶部署
+
+##### 1. 按照 `Langchain-Chatchat 体验部署` 内容顺序依次完成
+
+##### 2. 创建 chatchat 数据缓存路径
+```shell
+cd ~
+mkdir -p ~/chatchat
+```
+
+##### 3. 修改 `docker-compose.yaml` 文件内容
+
+原文件内容:
+```yaml
+  (上文 ...)
+  chatchat:
+    image: chatimage/chatchat:0.3.0-0622
+    (省略 ...)
+    # 将本地路径(~/chatchat/data)挂载到容器默认数据路径(/usr/local/lib/python3.11/site-packages/chatchat/data)中
+    # 将本地模型接入配置文件(~/chatchat/model_providers.yaml)挂载到容器默认模型接入配置文件路径(/usr/local/lib/python3.11/site-packages/chatchat/configs/)中
+    # volumes:
+    #   - ~/chatchat/data:/usr/local/lib/python3.11/site-packages/chatchat/data
+    #   - ~/chatchat/model_providers.yaml:/usr/local/lib/python3.11/site-packages/chatchat/configs/model_providers.yaml
+  (下文 ...)
+```
+将 `volumes` 字段注释打开, 并按照 `YAML` 格式对齐, 如下:
+```yaml
+  (上文 ...)
+  chatchat:
+    image: chatimage/chatchat:0.3.0-0622
+    (省略 ...)
+    # 将本地路径(~/chatchat/data)挂载到容器默认数据路径(/usr/local/lib/python3.11/site-packages/chatchat/data)中
+    # 将本地模型接入配置文件(~/chatchat/model_providers.yaml)挂载到容器默认模型接入配置文件路径(/usr/local/lib/python3.11/site-packages/chatchat/configs/)中
+    volumes:
+      - ~/chatchat/data:/usr/local/lib/python3.11/site-packages/chatchat/data
+      - ~/chatchat/model_providers.yaml:/usr/local/lib/python3.11/site-packages/chatchat/configs/model_providers.yaml
+  (下文 ...)
+```
+
+##### 4. 下载数据库初始文件
+
+> 提示: 这里的 `data.tar.gz` 文件仅包含初始化后的数据库 `samples` 文件一份及相应目录结构, 用户可将原先数据和目录结构迁移此处.
+> > [!WARNING] 请您先备份好您的数据再进行迁移!!!
+
+```shell
+cd ~/chatchat
+wget https://github.com/chatchat-space/Langchain-Chatchat/blob/master/docker/data.tar.gz
+tar -xvf data.tar.gz
+```
+```shell
+cd data
+pwd
+```
+/root/chatchat/data
+```shell
+ls -l
+```
+```text
+total 20
+drwxr-xr-x  3 root root 4096 Jun 22 10:46 knowledge_base
+drwxr-xr-x 18 root root 4096 Jun 22 10:52 logs
+drwxr-xr-x  5 root root 4096 Jun 22 10:46 media
+drwxr-xr-x  5 root root 4096 Jun 22 10:46 nltk_data
+drwxr-xr-x  3 root root 4096 Jun 22 10:46 temp
+```
+
+##### 5. 下载 `model_providers.yaml` 配置文件
+> 提示: 后续可以自定义本地路径下的 `model_providers.yaml` 来实现`自定义模型接入配置`
+```shell
+cd ~/chatchat
+wget https://github.com/chatchat-space/Langchain-Chatchat/blob/master/libs/model-providers/model_providers.yaml
+```
+
+##### 6. 重启 chatchat 服务
+
+这一步需要到 `docker-compose.yaml` 文件所在路径下执行, 即:
+```shell
+cd ~
+docker-compose down chatchat
+docker-compose up -d chatchat
+```
+操作及检查结果如下:
+```text
+[root@VM-2-15-centos ~]# docker-compose down chatchat
+WARN[0000] /root/docker-compose.yaml: `version` is obsolete 
+[+] Running 1/1
+ ✔ Container root-chatchat-1  Removed                                                                                               0.5s 
+[root@VM-2-15-centos ~]# docker-compose up -d
+WARN[0000] /root/docker-compose.yaml: `version` is obsolete 
+[+] Running 2/2
+ ✔ Container root-xinference-1  Running                                                                                             0.0s 
+ ✔ Container root-chatchat-1    Started                                                                                             0.2s
+[root@VM-2-15-centos ~]# docker-compose ps
+WARN[0000] /root/docker-compose.yaml: `version` is obsolete 
+NAME                IMAGE                           COMMAND                  SERVICE      CREATED          STATUS          PORTS
+root-chatchat-1     chatimage/chatchat:0.3.0-0622   "chatchat -a"            chatchat     33 seconds ago   Up 32 seconds   
+root-xinference-1   xprobe/xinference:v0.12.1       "/opt/nvidia/nvidia_…"   xinference   45 minutes ago   Up 45 minutes   
+[root@VM-2-15-centos ~]# ss -anptl | grep -E '(8501|7861|9997)'
+LISTEN 0      128          0.0.0.0:9997       0.0.0.0:*    users:(("pt_main_thread",pid=1489804,fd=21))
+LISTEN 0      128          0.0.0.0:8501       0.0.0.0:*    users:(("python",pid=1515944,fd=10))        
+LISTEN 0      128          0.0.0.0:7861       0.0.0.0:*    users:(("python",pid=1515878,fd=9))
+```