merges

Dockerfile

@@ -1,5 +1,5 @@
-# Use official Python image
-FROM python:3.10
+# Use official Python 3.11 image to match the wheel compatibility
+FROM python:3.11
 
 # Install system dependencies
 RUN apt-get update && apt-get install -y \
@@ -14,22 +14,26 @@ RUN apt-get update && apt-get install -y \
 # Set working directory
 WORKDIR /app
 
+# Set Hugging Face and Transformers cache to avoid permission errors
+ENV TRANSFORMERS_CACHE=/app/.cache \
+    HF_HOME=/app/.cache
+
+# Create and give full access to cache and model folders
+RUN mkdir -p /app/.cache /app/models && chmod -R 777 /app/.cache /app/models
+
 # Copy all project files
 COPY . .
 
-# Download and install llama-cpp-python prebuilt wheel from Hugging Face Dataset
-RUN wget https://huggingface.co/datasets/Kalpokoch/wheel-llama/resolve/main/llama_cpp_python-0.3.13-cp311-cp311-linux_x86_64.whl -O llama_cpp_python.whl && \
-    pip install llama_cpp_python.whl && \
-    rm llama_cpp_python.whl
+# ✅ Download and install llama-cpp-python wheel from Hugging Face Dataset
+RUN wget https://huggingface.co/datasets/Kalpokoch/wheel-llama/resolve/main/llama_cpp_python-0.3.13-cp311-cp311-linux_x86_64.whl && \
+    pip install llama_cpp_python-0.3.13-cp311-cp311-linux_x86_64.whl && \
+    rm llama_cpp_python-0.3.13-cp311-cp311-linux_x86_64.whl
 
 # Install remaining Python dependencies
 RUN pip install --no-cache-dir -r requirements.txt
 
-# Ensure /app/models exists and is writable
-RUN mkdir -p /app/models && chmod -R 777 /app/models
-
 # Expose FastAPI port
 EXPOSE 7860
 
-# Run the FastAPI app
+# Start FastAPI app
 CMD ["uvicorn", "app.app:app", "--host", "0.0.0.0", "--port", "7860"]

app/app.py

@@ -48,6 +48,10 @@ app.add_middleware(
     allow_headers=["*"],
 )
 
+@app.get("/")
+def read_root():
+    return {"message": "RAG chatbot backend is running!"}
+
 class ChatRequest(BaseModel):
     question: str
 

requirements.txt

@@ -1,6 +1,5 @@
 fastapi
 uvicorn
 huggingface_hub
-llama-cpp-python
 sentence-transformers
 scikit-learn