Update app.py

app.py

@@ -4,7 +4,6 @@ from huggingface_hub import InferenceClient
 # SEMANTIC SEARCH STEP 1
 from sentence_transformers import SentenceTransformer
 import torch
-#import lines go at the top!
 
 # SEMANTIC SEARCH STEP 2 --> EDIT WITH YOUR OWN KNOWLEDGE BASE WHEN READY
 with open("water_cycle.txt", "r", encoding="utf-8") as file:
@@ -27,13 +26,8 @@ def preprocess_text(text):
   for chunk in chunks:
     stripped_chunk = chunk.strip()
     cleaned_chunks.append(stripped_chunk)
-
-
-
-  # Print cleaned_chunks
   print(cleaned_chunks)
 
-  # Print the length of cleaned_chunks
   print(len(cleaned_chunks))
 
   # Return the cleaned_chunks
@@ -80,14 +74,10 @@ def get_top_chunks(query, chunk_embeddings, text_chunks):
   # Print the similarities
   print(similarities)
 
-
   # Find the indices of the 3 chunks with highest similarity scores
   top_indices = torch.topk(similarities, k=3).indices
-
-  # Print the top indices
   print(top_indices)
 
-
   # Create an empty list to store the most relevant chunks
   top_chunks = []
 
@@ -99,7 +89,6 @@ def get_top_chunks(query, chunk_embeddings, text_chunks):
   # Return the list of most relevant chunks
   return top_chunks
 
-
 # SEMANTIC SEARCH STEP 6
 
 # Call the get_top_chunks function with the original query
@@ -107,13 +96,13 @@ top_results = get_top_chunks('Is water good?',chunk_embeddings, cleaned_chunks)
 
 print(top_results)# Print the top results
 
-
+#the og code from gen ai lesson
 client = InferenceClient("microsoft/phi-4")
 # name of llm chatbot accessed ^^ or can use ' microsoft/phi-4 that's connected to the microsoft phi gen model
 
 def respond(message,history):
     
-    info = get_top_chunks(message, chunk_embeddings, text_chunks)
+    info = get_top_chunks(message, chunk_embeddings, cleaned_chunks)
     messages = [{'role': 'system','content':f'You are a friendly chatbot using {info} to answer questions.'}]
     #use string interporlation with variable info
 
@@ -127,14 +116,6 @@ def respond(message,history):
 
     return response['choices'][0]['message']['content'].strip()
 
-#def yes_or_no(message,history):
-#    return random.choice(['Yes','No'])
-    
-#def echo(message, history):
-    #always need two inputs
-#	return message
- 
-#print("Hello, World")
 
 chatbot = gr.ChatInterface(respond, type='messages')
 #defining my chatbot so user can interact, see their conversation and send new messages