Update Dockerfile

Dockerfile

@@ -1,4 +1,4 @@
-FROM nvidia/cuda:12.1.0-runtime-ubuntu22.04
+FROM nvidia/cuda:12.1.0-devel-ubuntu22.04
 
 # Install system dependencies
 RUN apt-get update && apt-get install -y \
@@ -6,21 +6,52 @@ RUN apt-get update && apt-get install -y \
     python3.10 \
     python3-pip \
     wget \
+    ninja-build \
+    gcc \
+    g++ \
     && rm -rf /var/lib/apt/lists/*
 
 WORKDIR /app
 
-# Install PyTorch first
-RUN pip3 install --no-cache-dir torch==2.4.0
+# Install basic Python packages first
+RUN pip3 install --no-cache-dir \
+    packaging \
+    setuptools \
+    wheel \
+    numpy \
+    torch==2.4.0
 
-# Install FastChat from source without editable mode
+# Install CUDA toolkit
+ENV CUDA_HOME=/usr/local/cuda
+ENV PATH=${CUDA_HOME}/bin:${PATH}
+ENV LD_LIBRARY_PATH=${CUDA_HOME}/lib64:${LD_LIBRARY_PATH}
+
+# Install dependencies in order
+COPY requirements.txt .
+RUN pip3 install --no-cache-dir \
+    transformers==4.43.2 \
+    accelerate \
+    peft \
+    datasets \
+    sentencepiece \
+    protobuf \
+    tiktoken \
+    scipy \
+    gradio \
+    cn2an>=0.5.22 \
+    langdetect>=1.0.9 \
+    openai \
+    tqdm \
+    && pip3 install --no-cache-dir flash-attn --no-build-isolation \
+    && pip3 install --no-cache-dir vllm==0.5.5 vllm-flash-attn
+
+# Install FastChat
 RUN git clone -b self-lengthen https://github.com/quanshr/FastChat.git && \
     cd FastChat && \
     pip3 install ".[model_worker,webui]"
 
-# Install other dependencies
-COPY requirements.txt .
-RUN pip3 install --no-cache-dir -r requirements.txt
+# Install LLaMA Factory
+RUN pip3 install --no-cache-dir llamafactory
 
 # Copy project files
 COPY . .