Update app.py

app.py

@@ -19,7 +19,7 @@ documents = [
     "Streamlit is an open-source Python library that makes it easy to build beautiful custom web-apps for machine learning and data science.",
     "Hugging Face is a company that provides tools and models for natural language processing (NLP).",
     "Retrieval-Augmented Generation (RAG) is a method that combines document retrieval with a generative model for question answering.",
-    "Letter of Recommendation August 29, 2024 Muthukumaran Azhagesan Senior Software Engineer +8326559594 I am Muthukumaran Azhagesan , a Senior Software Engineer and Lead at Cisco, USA, where I specialize in designing and developing enterprise software applications. With over two decades of experience in software engineering and enterprise security, I am the Founder, CEO, and CTO of Tame it AI...",
+    "Letter of Recommendation August 29, 2024 Muthukumaran Azhagesan Senior Software Engineer +8326559594 I am Muthukumaran Azhagesan, a Senior Software Engineer and Lead at Cisco...",
 ]
 
 # Encode the initial documents for similarity comparison
@@ -66,19 +66,24 @@ if st.button("💬 Get Answer"):
 
             # Step 3: Get the index of the most similar document
             top_index = similarity_scores[0].argmax()
-            retrieved_doc = documents[top_index]
+            
+            # Check if top_index is valid
+            if top_index < len(documents):
+                retrieved_doc = documents[top_index]
 
-            # Log the retrieved document for debugging
-            st.write(f"Retrieved Document: {retrieved_doc}")
+                # Log the retrieved document for debugging
+                st.write(f"Retrieved Document: {retrieved_doc}")
 
-            # Step 4: Use the Generator to Answer the Question
-            input_ids = generator_tokenizer.encode(f"question: {query} context: {retrieved_doc}", return_tensors="pt")
-            outputs = generator.generate(input_ids, max_length=200, num_return_sequences=1)
+                # Step 4: Use the Generator to Answer the Question
+                input_ids = generator_tokenizer.encode(f"question: {query} context: {retrieved_doc}", return_tensors="pt")
+                outputs = generator.generate(input_ids, max_length=200, num_return_sequences=1)
 
-            # Decode and display the response
-            answer = generator_tokenizer.decode(outputs[0], skip_special_tokens=True)
-            st.write("**Answer:**")
-            st.write(answer)
+                # Decode and display the response
+                answer = generator_tokenizer.decode(outputs[0], skip_special_tokens=True)
+                st.write("**Answer:**")
+                st.write(answer)
+            else:
+                st.error("No relevant document found.")
 
         except Exception as e:
             st.error(f"An error occurred: {str(e)}")