* build from cuda

Dockerfile

@@ -1,23 +1,37 @@
 # The data-juicer image includes all open-source contents of data-juicer,
 # and it will be instaled in editable mode.
 
-FROM python:3.10.15
+FROM nvidia/cuda:11.8.0-cudnn8-devel-ubuntu22.04
 
 # prepare the java env
 WORKDIR /opt
 # download jdk
-RUN wget https://aka.ms/download-jdk/microsoft-jdk-17.0.9-linux-x64.tar.gz -O jdk.tar.gz && \
-    tar -xzf jdk.tar.gz && \
-    rm -rf jdk.tar.gz && \
-    mv jdk-17.0.9+8 jdk
+RUN wget https://aka.ms/download-jdk/microsoft-jdk-17.0.9-linux-x64.tar.gz -O jdk.tar.gz \
+    && tar -xzf jdk.tar.gz \
+    && rm -rf jdk.tar.gz \
+    && mv jdk-17.0.9+8 jdk
 
 # set the environment variable
 ENV JAVA_HOME=/opt/jdk
 
+# install python 3.10
+RUN apt-get update \
+    && apt-get install -y git curl vim wget python3.10 libpython3.10-dev python3-pip \
+    && apt-get install -y libgl1-mesa-glx libglib2.0-0 \
+    && ln -sf /usr/bin/python3.10  /usr/bin/python3 \
+    && ln -sf /usr/bin/python3.10  /usr/bin/python \
+    && apt-get autoclean && rm -rf /var/lib/apt/lists/* \
+    && pip install --upgrade pip
+
+# install 3rd-party system dependencies
+RUN apt-get update \
+    && apt-get install ffmpeg libsm6 libxext6 software-properties-common build-essential cmake  -y
+
 WORKDIR /data-juicer
 
 # install requirements which need to be installed from source
-RUN pip install --upgrade setuptools && pip install git+https://github.com/xinyu1205/recognize-anything.git --default-timeout 1000
+RUN pip install --upgrade setuptools==69.5.1 setuptools_scm \
+    && pip install git+https://github.com/xinyu1205/recognize-anything.git --default-timeout 1000
 
 # install requirements first to better reuse installed library cache
 COPY environments/ environments/
@@ -27,6 +41,3 @@ RUN cat environments/* | grep -v '^#' | xargs pip install --default-timeout 1000
 COPY . .
 RUN pip install -v -e .[all]
 RUN pip install -v -e .[sandbox]
-
-# install 3rd-party system dependencies
-RUN apt-get update && apt-get install ffmpeg libsm6 libxext6  -y