Enable multiple workers to improve perf (#75)

* Add fast bm25 * Update * Fix bug * Fix bm25 bug * Fix bug * Refine code * Update multi-process * Add API to support upload local files (#67) * support upload file via API * add Readme for upload API * refactor query api * modify load_knowledge with session_config * use tempfile.mkdtemp() to store upload files * add docker image timezone for China (#68) * add image zone for China * remove unused ENV --------- Co-authored-by: shubao.sx <shubao.sx@alibaba-inc.com> Co-authored-by: Yue Fei <luxun.fy@alibaba-inc.com> * load data pipeline supports read config (#70) * Add gpu docker image timezone for China (#74) * Add fast bm25 (#66) * Add fast bm25 * Fix bm25 bug * Fix bug * Fix test * Update dockerfile * Fix bug * Update * Update docker file * Fix empty file bug * Fix local index error * Fix lint * Decouple gradio and backend * Add ui build * Add gunicorn * Fix gunicorn * Update nginx * add nginx image * Fix deployment issue * Fix upload --------- Co-authored-by: 筱文 <zxw320697@alibaba-inc.com> Co-authored-by: paradiseHIT <paradiseHIT@gmail.com> Co-authored-by: shubao.sx <shubao.sx@alibaba-inc.com>
aigc-apps · Jun 24, 2024 · 9e68ac6 · 9e68ac6
1 parent 7c2467e
commit 9e68ac6
Show file tree

Hide file tree

Showing 37 changed files with 2,140 additions and 602 deletions.
diff --git a/.github/workflows/docker.yml b/.github/workflows/docker.yml
@@ -52,3 +52,21 @@ jobs:
           docker tag ${{ env.REGISTRY }}/mybigpai/pairag:$IMAGE_TAG ${{ env.REGISTRY_HZ }}/mybigpai/pairag:$IMAGE_TAG
           docker push ${{ env.REGISTRY }}/mybigpai/pairag:$IMAGE_TAG
           docker push ${{ env.REGISTRY_HZ }}/mybigpai/pairag:$IMAGE_TAG
+
+      - name: Build and push UI image
+        env:
+          IMAGE_TAG: 0.0.2_ui
+        run: |
+          docker build -t ${{ env.REGISTRY }}/mybigpai/pairag:$IMAGE_TAG -f Dockerfile_ui .
+          docker tag ${{ env.REGISTRY }}/mybigpai/pairag:$IMAGE_TAG ${{ env.REGISTRY_HZ }}/mybigpai/pairag:$IMAGE_TAG
+          docker push ${{ env.REGISTRY }}/mybigpai/pairag:$IMAGE_TAG
+          docker push ${{ env.REGISTRY_HZ }}/mybigpai/pairag:$IMAGE_TAG
+
+      - name: Build and push nginx image
+        env:
+          IMAGE_TAG: 0.0.2_nginx
+        run: |
+          docker build -t ${{ env.REGISTRY }}/mybigpai/pairag:$IMAGE_TAG -f Dockerfile_nginx .
+          docker tag ${{ env.REGISTRY }}/mybigpai/pairag:$IMAGE_TAG ${{ env.REGISTRY_HZ }}/mybigpai/pairag:$IMAGE_TAG
+          docker push ${{ env.REGISTRY }}/mybigpai/pairag:$IMAGE_TAG
+          docker push ${{ env.REGISTRY_HZ }}/mybigpai/pairag:$IMAGE_TAG
diff --git a/.gitignore b/.gitignore
@@ -222,3 +222,4 @@ output
 *.local.toml
 
 localdata/
+*.tmp
diff --git a/Dockerfile b/Dockerfile
@@ -24,4 +24,4 @@ RUN apt-get update && apt-get install -y libgl1 libglib2.0-0
 WORKDIR /app
 COPY . .
 COPY --from=builder ${VIRTUAL_ENV} ${VIRTUAL_ENV}
-ENTRYPOINT ["pai_rag", "run"]
+CMD ["pai_rag", "run"]
diff --git a/Dockerfile_gpu b/Dockerfile_gpu
@@ -26,4 +26,4 @@ RUN apt-get update && apt-get install -y libgl1 libglib2.0-0
 WORKDIR /app
 COPY . .
 COPY --from=builder ${VIRTUAL_ENV} ${VIRTUAL_ENV}
-ENTRYPOINT ["pai_rag", "run"]
+CMD ["pai_rag", "run"]
diff --git a/Dockerfile_nginx b/Dockerfile_nginx
@@ -0,0 +1,3 @@
+FROM nginx:latest
+COPY ./nginx/default.conf etc/nginx/conf.d/default.conf
+COPY ./nginx/nginx.conf etc/nginx/nginx.conf
diff --git a/Dockerfile_ui b/Dockerfile_ui
@@ -0,0 +1,27 @@
+FROM python:3.10-slim AS builder
+
+RUN pip3 install poetry
+
+ENV POETRY_NO_INTERACTION=1 \
+    POETRY_VIRTUALENVS_IN_PROJECT=1 \
+    POETRY_VIRTUALENVS_CREATE=1 \
+    POETRY_CACHE_DIR=/tmp/poetry_cache
+
+WORKDIR /app
+COPY . .
+
+RUN poetry install && rm -rf $POETRY_CACHE_DIR
+
+FROM python:3.10-slim AS prod
+
+RUN rm -rf /etc/localtime && ln -s /usr/share/zoneinfo/Asia/Harbin  /etc/localtime
+
+ENV VIRTUAL_ENV=/app/.venv \
+    PATH="/app/.venv/bin:$PATH"
+
+RUN apt-get update && apt-get install -y libgl1 libglib2.0-0
+
+WORKDIR /app
+COPY . .
+COPY --from=builder ${VIRTUAL_ENV} ${VIRTUAL_ENV}
+CMD ["pai_rag", "ui"]
diff --git a/README.md b/README.md
@@ -36,19 +36,19 @@ poetry install
 load_data -c src/pai_rag/config/settings.yaml -d directory_path
 ```
 
-#### Step4: 启动程序
+#### Step4: 启动RAG服务
 
 使用OpenAI API，需要在命令行引入环境变量 export OPENAI_API_KEY=""
 使用DashScope API，需要在命令行引入环境变量 export DASHSCOPE_API_KEY=""
 
 ```bash
-# 启动，支持自定义host(默认0.0.0.0), port(默认8000), config(默认src/pai_rag/config/settings.yaml)
-pai_rag run [--host HOST] [--port PORT] [--config CONFIG_FILE]
+# 启动，支持自定义host(默认0.0.0.0), port(默认8001), config(默认src/pai_rag/config/settings.yaml)
+pai_rag serve [--host HOST] [--port PORT] [--config CONFIG_FILE]
 ```
 
-现在你可以使用命令行向服务侧发送API请求，或者直接打开http://localhost:8000
+你可以使用命令行向服务侧发送API请求。比如调用[Upload API](#upload-api)上传知识库文件。
 
-1. 对话
+##### Query API
 
 - **Rag Query请求**
 
@@ -77,7 +77,7 @@ curl -X 'POST' http://127.0.0.1:8000/service/query -H "Content-Type: application
 curl -X 'POST' http://127.0.0.1:8000/service/query/agent -H "Content-Type: application/json" -d '{"question":"今年是2024年，10年前是哪一年？"}'
 ```
 
-2. 评估
+##### Evaluation API
 
 支持三种评估模式：全链路评估、检索效果评估、生成效果评估。
 
@@ -144,7 +144,7 @@ curl -X 'POST' http://127.0.0.1:8000/service/batch_evaluate/response
 }
 ```
 
-3. 上传
+##### Upload API
 
 支持通过API的方式上传本地文件，并支持指定不同的faiss_path，每次发送API请求会返回一个task_id，之后可以通过task_id来查看文件上传状态（processing、completed、failed）。
 
@@ -164,6 +164,15 @@ curl http://127.0.0.1:8077/service/get_upload_state\?task_id\=2c1e557733764fdb9f
 # Return: {"task_id":"2c1e557733764fdb9fefa063538914da","status":"completed"}
 ```
 
+### RAG WEB UI
+
+```bash
+# 启动，支持自定义host(默认0.0.0.0), port(默认8002), config(默认localhost:8001)
+pai_rag ui [--host HOST] [--port PORT] [rag-url RAG_URL]
+```
+
+你也可以打开http://127.0.0.1:8002/ 来配置RAG服务以及上传本地数据。
+
 ### 独立脚本文件：不依赖于整体服务的启动，可独立运行
 
 1. 向当前索引存储中插入新文件

diff --git a/nginx/default.conf b/nginx/default.conf
@@ -0,0 +1,67 @@
+
+server {
+    listen       8000;
+    listen  [::]:8000;
+    server_name  localhost;
+    client_max_body_size 50m;
+
+    #access_log  /var/log/nginx/host.access.log  main;
+
+    location / {
+        proxy_set_header Host \$host;
+        proxy_set_header X-Forwarded-For \$proxy_add_x_forwarded_for;
+        proxy_pass http://127.0.0.1:8002;
+    }
+
+    #Websocket configuration
+    location /queue/ {
+    proxy_pass http://127.0.0.1:8002/queue/;
+    proxy_http_version 1.1;
+    proxy_set_header Upgrade $http_upgrade;
+    proxy_set_header Connection "upgrade";
+    }
+
+    location /service {
+        proxy_set_header Host \$host;
+        proxy_set_header X-Forwarded-For \$proxy_add_x_forwarded_for;
+        proxy_pass http://127.0.0.1:8001;
+    }
+
+    location /docs {
+        proxy_set_header Host \$host;
+        proxy_set_header X-Forwarded-For \$proxy_add_x_forwarded_for;
+        proxy_pass http://127.0.0.1:8001;
+    }
+
+    #error_page  404              /404.html;
+
+    # redirect server error pages to the static page /50x.html
+    #
+    error_page   500 502 503 504  /50x.html;
+    location = /50x.html {
+        root   /usr/share/nginx/html;
+    }
+
+    # proxy the PHP scripts to Apache listening on 127.0.0.1:80
+    #
+    #location ~ \.php$ {
+    #    proxy_pass   http://127.0.0.1;
+    #}
+
+    # pass the PHP scripts to FastCGI server listening on 127.0.0.1:9000
+    #
+    #location ~ \.php$ {
+    #    root           html;
+    #    fastcgi_pass   127.0.0.1:9000;
+    #    fastcgi_index  index.php;
+    #    fastcgi_param  SCRIPT_FILENAME  /scripts$fastcgi_script_name;
+    #    include        fastcgi_params;
+    #}
+
+    # deny access to .htaccess files, if Apache's document root
+    # concurs with nginx's one
+    #
+    #location ~ /\.ht {
+    #    deny  all;
+    #}
+}
diff --git a/nginx/nginx.conf b/nginx/nginx.conf
@@ -0,0 +1,33 @@
+
+user  nginx;
+daemon off;
+worker_processes  auto;
+
+error_log  /var/log/nginx/error.log notice;
+pid        /var/run/nginx.pid;
+
+
+events {
+    worker_connections  1024;
+}
+
+
+http {
+    include       /etc/nginx/mime.types;
+    default_type  application/octet-stream;
+
+    log_format  main  '$remote_addr - $remote_user [$time_local] "$request" '
+                      '$status $body_bytes_sent "$http_referer" '
+                      '"$http_user_agent" "$http_x_forwarded_for"';
+
+    access_log  /var/log/nginx/access.log  main;
+
+    sendfile        on;
+    #tcp_nopush     on;
+
+    keepalive_timeout  65;
+
+    #gzip  on;
+
+    include /etc/nginx/conf.d/*.conf;
+}
Original file line number	Diff line number	Diff line change
Expand Up		@@ -222,3 +222,4 @@ output
		*.local.toml

		localdata/
		*.tmp