Update app.py

app.py

@@ -306,6 +306,7 @@ def process_news(query, temperature, top_p, repetition_penalty, news_source):
             
             full_summary, cleaned_summary = summarize_news_content(clean_content, model)
             relevance_score = calculate_relevance_score(cleaned_summary, model)
+            print(f"Relevance score for article '{article['title']}': {relevance_score}")  # Debug print
             
             processed_article = {
                 "published_date": article["published_date"],
@@ -314,11 +315,12 @@ def process_news(query, temperature, top_p, repetition_penalty, news_source):
                 "content": clean_content,
                 "summary": full_summary,
                 "cleaned_summary": cleaned_summary,
-                "relevance_score": relevance_score
+                "relevance_score": relevance_score  # Ensure this line is present
             }
             processed_articles.append(processed_article)
         except Exception as e:
             print(f"Error processing article: {str(e)}")
+    
         
     if not processed_articles:
         return f"Failed to process any news articles from {news_source}. Please try again or check the summarization process."
@@ -455,10 +457,19 @@ def export_news_to_excel():
     df['summary'] = df['cleaned_summary']
     df = df.drop(columns=['cleaned_summary'])  # Remove the extra column
     
+    # Ensure relevance_score is present and convert to float
+    if 'relevance_score' not in df.columns:
+        df['relevance_score'] = 0.0
+    else:
+        df['relevance_score'] = df['relevance_score'].astype(float)
+    
     # Reorder columns to put relevance_score after summary
     columns = ['published_date', 'title', 'url', 'content', 'summary', 'relevance_score']
     df = df[columns]
     
+    # Format relevance_score to display 2 decimal places
+    df['relevance_score'] = df['relevance_score'].apply(lambda x: f"{x:.2f}")
+    
     with NamedTemporaryFile(delete=False, suffix='.xlsx') as tmp:
         excel_path = tmp.name
         df.to_excel(excel_path, index=False)
@@ -487,13 +498,17 @@ def calculate_relevance_score(summary, model):
     chain = LLMChain(llm=model, prompt=prompt_template)
     response = chain.run(summary=summary)
     
+    print(f"Raw relevance score response: {response}")  # Debug print
+    
     try:
         score = float(response.strip())
-        return min(max(score, 0.00), 1.00)  # Ensure the score is between 0.00 and 1.00
+        final_score = min(max(score, 0.00), 1.00)  # Ensure the score is between 0.00 and 1.00
+        print(f"Processed relevance score: {final_score}")  # Debug print
+        return final_score
     except ValueError:
         print(f"Error parsing relevance score: {response}")
         return 0.00
-        
+    
 def ask_question(question, temperature, top_p, repetition_penalty, web_search, google_news_rss):
     global conversation_history