Update Dockerfile

Dockerfile

@@ -2,33 +2,32 @@
 FROM python:3.10-slim-buster
 
 # Set the working directory in the container
-WORKDIR /app  # Changed from /app/backend to /app
+WORKDIR /app
 
-# Install system dependencies needed for MLC LLM
+# Install system dependencies needed for MLC LLM (e.g., git, cmake for build tools if compiling on the fly)
+# These are necessary for MLC LLM's build process if models are compiled within the container
 RUN apt-get update && apt-get install -y --no-install-recommends \
     git \
     build-essential \
     cmake \
-    git-lfs \
-    ffmpeg \
-    libsm6 \
-    libxext6 \
-    rsync \
-    libgl1 \
-    && rm -rf /var/lib/apt/lists/* \
-    && git lfs install
+    && rm -rf /var/lib/apt/lists/*
 
 # Copy the requirements file and install Python dependencies
 COPY requirements.txt .
-RUN pip install --no-cache-dir -r requirements.txt || echo "Failed to install requirements"
+RUN pip install --no-cache-dir -r requirements.txt
 
-# Install torch specifically for CUDA if using GPU
+# Install torch specifically for CUDA if using GPU, otherwise use CPU
+# For HuggingFace Spaces with GPU, CUDA 11.8 is a common and recommended version.
+# Users might need to adjust 'cu118' to 'cpu' or a different CUDA version based on their target hardware.
+# MLC-LLM might have its own torch dependency, ensure compatibility or remove this line if MLC-LLM handles it.
 RUN pip uninstall -y torch && pip install torch==2.1.0 --extra-index-url https://download.pytorch.org/whl/cu118
 
 # Copy the Flask application files
 COPY app.py .
 
-# Copy the model artifacts
+# Copy the model artifacts. This assumes model_artifacts exists and is populated.
+# For large models, consider using git-lfs for HuggingFace Spaces or downloading at runtime
+# if the model is too large for the Docker image or needs dynamic loading.
 COPY model_artifacts ./model_artifacts
 
 # Expose the port the app runs on
@@ -36,7 +35,9 @@ EXPOSE 5000
 
 # Define environment variables for MLC LLM model paths
 ENV MLC_MODEL_ARTIFACTS_DIR="./model_artifacts"
-ENV MLC_MODEL_NAME="Llama-2-7b-chat-hf-q4f16_1"
+ENV MLC_MODEL_NAME="Llama-2-7b-chat-hf-q4f16_1" # Ensure this matches your downloaded model
 
 # Command to run the Flask application
-CMD ["python", "app.py"]
+# Using Flask's built-in server for simplicity in development and small deployments.
+# For production-grade deployments, consider a WSGI server like Gunicorn (e.g., gunicorn --bind 0.0.0.0:5000 app:app)
+CMD ["python", "app.py"]