Update Dockerfile

Dockerfile

@@ -1,31 +1,90 @@
-# Create Dockerfile
-FROM python:3.10-slim
+# Multi-stage build: Train models then run app
+
+# ============================================
+# Stage 1: Training Stage
+# ============================================
+FROM python:3.10-slim as trainer
 
 WORKDIR /app
 
-# Install system dependencies
+# Install system dependencies for training
 RUN apt-get update && apt-get install -y \
     git \
+    build-essential \
     && rm -rf /var/lib/apt/lists/*
 
-# Copy requirements first for better caching
+# Copy requirements and install
 COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy only necessary files for training
+COPY models/ models/
+COPY data/ data/
+
+# Create output directory
+RUN mkdir -p models/model_weights/custom_models
 
-# Install Python dependencies
+# Train models
+RUN echo "🔄 Starting model training..." && \
+    python models/train_model.py && \
+    echo "✅ Training completed!"
+
+# Verify models were created and have predict_proba
+RUN echo "🔍 Verifying models..." && \
+    ls -lah models/model_weights/custom_models/ && \
+    python -c "import joblib; \
+    m_en = joblib.load('models/model_weights/custom_models/english_model.pkl'); \
+    m_bn = joblib.load('models/model_weights/custom_models/bengali_model.pkl'); \
+    print(f'✅ English Model: {type(m_en).__name__}'); \
+    print(f'✅ Has predict_proba: {hasattr(m_en, \"predict_proba\")}'); \
+    print(f'✅ Bengali Model: {type(m_bn).__name__}'); \
+    print(f'✅ Has predict_proba: {hasattr(m_bn, \"predict_proba\")}')"
+
+# ============================================
+# Stage 2: Runtime Stage
+# ============================================
+FROM python:3.10-slim
+
+WORKDIR /app
+
+# Install runtime system dependencies
+RUN apt-get update && apt-get install -y \
+    git \
+    && rm -rf /var/lib/apt/lists/*
+
+# Copy requirements and install
+COPY requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt
 
-# Copy application files
+# Copy application code
 COPY . .
 
-# Create cache directories
+# Copy trained models from training stage
+COPY --from=trainer /app/models/model_weights/custom_models/*.pkl models/model_weights/custom_models/
+COPY --from=trainer /app/models/model_weights/custom_models/metadata.json models/model_weights/custom_models/
+
+# Create cache directories for transformers
 RUN mkdir -p /tmp/transformers_cache /tmp/huggingface
 
 # Set environment variables
 ENV TRANSFORMERS_CACHE=/tmp/transformers_cache
 ENV HF_HOME=/tmp/huggingface
+ENV PYTHONUNBUFFERED=1
+
+# Verify models exist in final image
+RUN echo "🔍 Final verification..." && \
+    ls -lah models/model_weights/custom_models/ && \
+    python -c "import joblib; \
+    m = joblib.load('models/model_weights/custom_models/english_model.pkl'); \
+    print(f'✅ Final Stage - Model loaded: {type(m).__name__}'); \
+    print(f'✅ Has predict_proba: {hasattr(m, \"predict_proba\")}')"
 
 # Expose port 7860 (Hugging Face Spaces default)
 EXPOSE 7860
 
+# Health check
+HEALTHCHECK --interval=30s --timeout=10s --start-period=60s --retries=3 \
+    CMD python -c "import requests; requests.get('http://localhost:7860/health', timeout=5)" || exit 1
+
 # Run the application
-CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]
+CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860", "--workers", "1"]