Initial commit for Hugging Face Space deployment

Dockerfile

@@ -14,14 +14,26 @@ COPY requirements.txt .
 # Install Python dependencies
 RUN pip install --no-cache-dir -r requirements.txt
 
+# Create cache directories with proper permissions
+RUN mkdir -p /.cache && chmod 777 /.cache
+RUN mkdir -p /root/.cache && chmod 777 /root/.cache
+RUN mkdir -p /app/.cache && chmod 777 /app/.cache
+
+# Create models directory for Hugging Face
+RUN mkdir -p /app/models && chmod 777 /app/models
+ENV TRANSFORMERS_CACHE=/app/models
+
 # Copy the rest of the application
 COPY . .
 
-# Create necessary directories
-RUN mkdir -p data/documents data/vector_db
+# Create necessary directories with proper permissions
+RUN mkdir -p data/documents data/vector_db && \
+    chmod -R 777 data
 
 # Set environment variable to avoid TOKENIZERS_PARALLELISM warning
 ENV TOKENIZERS_PARALLELISM=false
+ENV HF_HOME=/app/.cache
+ENV XDG_CACHE_HOME=/app/.cache
 
 # Expose the port required by Hugging Face Spaces
 EXPOSE 7860

app/core/llm.py

@@ -28,9 +28,20 @@ def get_llm():
 
 def get_embeddings():
     """Initialize and return the embeddings model."""
+    # Set up cache directories with proper permissions
+    cache_dir = "/app/models"
+    if not os.path.exists(cache_dir):
+        try:
+            os.makedirs(cache_dir, exist_ok=True)
+            os.chmod(cache_dir, 0o777)
+        except Exception as e:
+            print(f"Warning: Could not create cache directory: {e}")
+            cache_dir = None
+    
     # SentenceTransformers can be used locally without an API key
     return HuggingFaceEmbeddings(
-        model_name=EMBEDDING_MODEL
+        model_name=EMBEDDING_MODEL,
+        cache_folder=cache_dir
     )
 
 def get_chat_model():