updated dockerfile

Dockerfile

@@ -1,6 +1,6 @@
 FROM python:3.11-slim
 
-# Install required system dependencies needed for llama-cpp-python compilation
+# Install required system dependencies
 RUN apt-get update && apt-get install -y \
     git curl build-essential cmake \
     && rm -rf /var/lib/apt/lists/*
@@ -8,29 +8,36 @@ RUN apt-get update && apt-get install -y \
 # Set working directory
 WORKDIR /app
 
-# Create writable directories for cache and the persistent vector DB
-# Note: For production, consider using a non-root user and more specific permissions
+# Create writable directories
 RUN mkdir -p /app/.cache /app/vector_database && chmod -R 777 /app
 
-# Set environment variables for huggingface cache and to disable chroma telemetry
+# Set environment variables
 ENV TRANSFORMERS_CACHE=/app/.cache \
     HF_HOME=/app/.cache \
     CHROMADB_DISABLE_TELEMETRY=true
 
-# ✅ RECOMMENDATION: To avoid version conflicts, it's best to remove 'llama-cpp-python' 
-# from your requirements.txt and rely on this explicit, version-pinned installation.
-RUN pip install --no-cache-dir llama-cpp-python==0.2.61
-
-# Install other dependencies from requirements.txt
+# Install dependencies from requirements.txt first
 COPY requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt
 
-# Copy the application code and the processed data file
+# ✅ STEP 1: Copy the source data and the Python script into the image
+COPY ./combined_context.jsonl .
+COPY ./create_granular_chunks.py .
+
+# ✅ STEP 2: Run the script to generate the chunks file inside the image
+RUN python create_granular_chunks.py
+
+# ✅ STEP 3: The 'granular_chunks_improved.jsonl' now exists inside the image.
+# We no longer need to copy it from our local machine.
+
+# Note: As recommended before, 'llama-cpp-python' should be removed from requirements.txt
+# to rely on the more stable, version-pinned installation below.
+RUN pip install --no-cache-dir llama-cpp-python==0.2.61
+
+# Copy the rest of the application code
 COPY ./app ./app
-# ✅ CORRECTED FILENAME: Ensure this matches the output of your chunking script
-COPY ./granular_chunks_improved.jsonl .
 
-# Download your fine-tuned TinyLlama GGUF model from Hugging Face
+# Download your fine-tuned TinyLlama GGUF model
 RUN curl -fL -o /app/tinyllama_dop_q4_k_m.gguf \
     https://huggingface.co/Kalpokoch/FinetunedQuantizedTinyLama/resolve/main/tinyllama_dop_q4_k_m.gguf \
     && echo "✅ TinyLlama model downloaded."
@@ -38,5 +45,5 @@ RUN curl -fL -o /app/tinyllama_dop_q4_k_m.gguf \
 # Expose the application port
 EXPOSE 7860
 
-# Run the FastAPI application using uvicorn
+# Run the FastAPI application
 CMD ["uvicorn", "app.app:app", "--host", "0.0.0.0", "--port", "7860"]