feat: Support 8-bit quantization and 4-bit quantization for multi-gpu inference

docker/base/Dockerfile

@@ -1,25 +1,48 @@
-FROM nvidia/cuda:11.8.0-devel-ubuntu22.04
+ARG BASE_IMAGE="nvidia/cuda:11.8.0-devel-ubuntu22.04"
+
+FROM ${BASE_IMAGE}
+ARG BASE_IMAGE
 
 RUN apt-get update && apt-get install -y git python3 pip wget \
     && apt-get clean
 
-# download code from githu: https://github.com/csunny/DB-GPT
-# ENV DBGPT_VERSION="v0.3.3"
-# RUN wget https://github.com/csunny/DB-GPT/archive/refs/tags/$DBGPT_VERSION.zip
+ARG BUILD_LOCAL_CODE="false"
+ARG LANGUAGE="en"
+ARG PIP_INDEX_URL="https://pypi.org/simple"
+ENV PIP_INDEX_URL=$PIP_INDEX_URL
 
-# clone latest code, and rename to /app
-RUN git clone https://github.com/csunny/DB-GPT.git /app
+# COPY only requirements.txt first to leverage Docker cache
+COPY ./requirements.txt /tmp/requirements.txt
 
 WORKDIR /app
 
-RUN pip3 install --upgrade pip \
-    && pip3 install --no-cache-dir -r requirements.txt \
-    && pip3 install seaborn mpld3 \
-    && wget https://github.com/explosion/spacy-models/releases/download/zh_core_web_sm-3.5.0/zh_core_web_sm-3.5.0-py3-none-any.whl -O /tmp/zh_core_web_sm-3.5.0-py3-none-any.whl \
-    && pip3 install /tmp/zh_core_web_sm-3.5.0-py3-none-any.whl \
-    && rm /tmp/zh_core_web_sm-3.5.0-py3-none-any.whl \
-    && rm -rf `pip3 cache dir` 
+RUN pip3 install --upgrade pip -i $PIP_INDEX_URL \
+    && (if [ "${BUILD_LOCAL_CODE}" = "false" ]; \
+        # if not build local code, clone latest code from git, and rename to /app, TODO: download by version, like: https://github.com/eosphoros-ai/DB-GPT/archive/refs/tags/$DBGPT_VERSION.zip
+        then git clone https://github.com/eosphoros-ai/DB-GPT.git /app \
+        && cp /app/requirements.txt /tmp/requirements.txt; \
+    fi;) \
+    && pip3 install -r /tmp/requirements.txt -i $PIP_INDEX_URL --no-cache-dir \
+    && rm /tmp/requirements.txt
 
-# RUN python3 -m spacy download zh_core_web_sm
+RUN (if [ "${LANGUAGE}" = "zh" ]; \
+        # language is zh, download zh_core_web_sm from github
+        then wget https://github.com/explosion/spacy-models/releases/download/zh_core_web_sm-3.5.0/zh_core_web_sm-3.5.0-py3-none-any.whl -O /tmp/zh_core_web_sm-3.5.0-py3-none-any.whl \
+        && pip3 install /tmp/zh_core_web_sm-3.5.0-py3-none-any.whl -i $PIP_INDEX_URL \
+        && rm /tmp/zh_core_web_sm-3.5.0-py3-none-any.whl; \
+        # not zh, download directly
+        else python3 -m spacy download zh_core_web_sm; \
+    fi;) \
+    && rm -rf `pip3 cache dir`
+
+ARG BUILD_LOCAL_CODE="false"
+# COPY the rest of the app
+COPY . /tmp/app
+
+# TODO：Need to find a better way to determine whether to build docker image with local code.
+RUN (if [ "${BUILD_LOCAL_CODE}" = "true" ]; \
+    then mv /tmp/app / && rm -rf /app/logs && rm -rf /app/pilot/data && rm -rf /app/pilot/message; \
+    else rm -rf /tmp/app; \
+    fi;)
 
 EXPOSE 5000

docker/base/build_image.sh

@@ -4,5 +4,72 @@ SCRIPT_LOCATION=$0
 cd "$(dirname "$SCRIPT_LOCATION")"
 WORK_DIR=$(pwd)
 
+BASE_IMAGE="nvidia/cuda:11.8.0-devel-ubuntu22.04"
 IMAGE_NAME="db-gpt"
-docker build -f Dockerfile -t $IMAGE_NAME $WORK_DIR/../../
+# zh: https://pypi.tuna.tsinghua.edu.cn/simple
+PIP_INDEX_URL="https://pypi.org/simple"
+# en or zh
+LANGUAGE="en"
+BUILD_LOCAL_CODE="false"
+
+usage () {
+    echo "USAGE: $0 [--base-image nvidia/cuda:11.8.0-devel-ubuntu22.04] [--image-name db-gpt]"
+    echo "  [-b|--base-image base image name] Base image name"
+    echo "  [-n|--image-name image name] Current image name, default: db-gpt"
+    echo "  [-i|--pip-index-url pip index url] Pip index url, default: https://pypi.org/simple"
+    echo "  [--language en or zh] You language, default: en"
+    echo "  [--build-local-code true or false] Whether to use the local project code to package the image, default: false"
+    echo "  [-h|--help] Usage message"
+}
+
+while [[ $# -gt 0 ]]; do
+    key="$1"
+    case $key in
+        -b|--base-image)
+        BASE_IMAGE="$2"
+        shift # past argument
+        shift # past value
+        ;;
+        -n|--image-name)
+        IMAGE_NAME="$2"
+        shift # past argument
+        shift # past value
+        ;;
+        -i|--pip-index-url)
+        PIP_INDEX="$2"
+        shift
+        shift
+        ;;
+        --language)
+        LANGUAGE="$2"
+        shift
+        shift
+        ;;
+        --build-local-code)
+        BUILD_LOCAL_CODE="$2"
+        shift
+        shift
+        ;;
+        -h|--help)
+        help="true"
+        shift
+        ;;
+        *)
+        usage
+        exit 1
+        ;;
+    esac
+done
+
+if [[ $help ]]; then
+    usage
+    exit 0
+fi
+
+docker build \
+    --build-arg BASE_IMAGE=$BASE_IMAGE \
+    --build-arg PIP_INDEX_URL=$PIP_INDEX_URL \
+    --build-arg LANGUAGE=$LANGUAGE \
+    --build-arg BUILD_LOCAL_CODE=$BUILD_LOCAL_CODE \
+    -f Dockerfile \
+    -t $IMAGE_NAME $WORK_DIR/../../

docker/build_all_images.sh

@@ -4,6 +4,11 @@ SCRIPT_LOCATION=$0
 cd "$(dirname "$SCRIPT_LOCATION")"
 WORK_DIR=$(pwd)
 
-bash $WORK_DIR/base/build_image.sh
+bash $WORK_DIR/base/build_image.sh "$@"
+
+if [ 0 -ne $? ]; then
+    ehco "Error: build base image failed"
+    exit 1
+fi
 
 bash $WORK_DIR/allinone/build_image.sh