Synced repo using 'sync_with_huggingface' Github Action

Dockerfile

@@ -18,6 +18,8 @@ RUN pip install --no-cache-dir -r requirements.txt
 COPY . .
 
 RUN chmod +x /app/entrypoint.sh
+RUN mkdir -p /app/data && chmod -R 777 /app/data
+
 
 EXPOSE 8501
 

scrape_chroma.py

@@ -43,7 +43,8 @@ def scrape_and_store():
     print(f"Fetched {len(all_titles_sources)} unique titles.")
 
     # Save to CSV
-    save_titles_to_csv(all_titles_sources, filename="pib_titles.csv")
+    save_titles_to_csv(all_titles_sources, filename="data/pib_titles.csv")
+
 
     # Prepare for ChromaDB
     documents = [title for title, source in all_titles_sources]