Update Dockerfile

Dockerfile

@@ -12,22 +12,41 @@ RUN apt-get update && apt-get install -y wget ca-certificates && \
       rm libssl1.1_1.1.1f-1ubuntu2_amd64.deb)) && \
     apt-get clean && rm -rf /var/lib/apt/lists/*
 
+# 安装PyTorch CPU版本
 RUN pip install torch==2.1.1 -i https://download.pytorch.org/whl/cpu
-RUN pip install paddlepaddle==2.5.1 -i https://mirror.baidu.com/pypi/simple
-RUN pip install transformers gradio Pillow fastapi
-RUN pip install paddleocr==2.7.0.3 \
+
+# 直接升级到最新的PaddlePaddle 3.0
+RUN pip install paddlepaddle==3.0.0 || \
+    pip install paddlepaddle -i https://mirror.baidu.com/pypi/simple
+
+# 安装基础依赖
+RUN pip install transformers gradio Pillow fastapi uvicorn
+
+# 直接升级到PaddleOCR 3.0最新版
+RUN pip install paddleocr==3.0.3 \
     && pip uninstall -y opencv-python opencv-contrib-python \
     && pip install opencv-python-headless
 
+# PaddleOCR 3.0额外依赖
+RUN pip install shapely pyclipper
+
+# 创建用户
 RUN useradd -m -u 1000 user
 USER user
 ENV HOME=/home/user
 WORKDIR $HOME/app
-RUN mkdir output
+
+# 创建输出目录
+RUN mkdir -p output
+
+# 复制文件
 COPY --chown=user app.py $HOME/app
 COPY --chown=user table.jpg $HOME/app 
 COPY --chown=user word_1.jpg $HOME/app
-RUN paddleocr --image_dir ./word_1.jpg --use_angle_cls true --use_gpu false  --lang ch
-RUN paddleocr --image_dir ./table.jpg --type=structure --table=true  --lang ch
 
+# 使用PaddleOCR 3.0的新CLI命令预热模型
+RUN paddleocr ocr -i ./word_1.jpg --lang ch --device cpu
+RUN paddleocr table_structure -i ./table.jpg --lang ch --device cpu || echo "Table structure preload completed"
+
+# 启动命令
 CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]