Update Dockerfile

Dockerfile

@@ -1,39 +1,30 @@
-# Use NVIDIA CUDA base image
 FROM nvidia/cuda:12.1.1-base-ubuntu22.04
 
-# Install system dependencies
+# 1. Install system dependencies properly
 RUN apt-get update && apt-get install -y \
     python3.10 \
     python3-pip \
+    python3.10-venv \
     git \
     && rm -rf /var/lib/apt/lists/*
 
-# Create a non-root user and working directory
-RUN useradd -m appuser && \
-    mkdir -p /app && \
-    chown appuser:appuser /app
+# 2. Create and activate virtual environment
+RUN python3.10 -m venv /opt/venv
+ENV PATH="/opt/venv/bin:$PATH"
 
-# Set environment variables for Hugging Face cache
+# 3. Set cache directories (writable locations)
 ENV TRANSFORMERS_CACHE=/app/model_cache \
     HF_HOME=/app/huggingface \
     XDG_CACHE_HOME=/app/cache
 
-# Switch to non-root user
-USER appuser
 WORKDIR /app
 
-# Create cache directories
-RUN mkdir -p $TRANSFORMERS_CACHE $HF_HOME $XDG_CACHE_HOME
-
-# Install Python dependencies
-COPY --chown=appuser:appuser requirements.txt .
+# 4. Install requirements FIRST (better layer caching)
+COPY requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt
 
-# Copy application code
-COPY --chown=appuser:appuser . .
-
-# Expose port (7860 in your error, adjust if needed)
-EXPOSE 7860
+# 5. Copy app code
+COPY . .
 
-# Run FastAPI
-CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]
+# 6. Explicitly use the virtualenv's uvicorn
+CMD ["/opt/venv/bin/uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]