Update app.py

app.py

@@ -88,7 +88,7 @@ def generate_tokens(model, generator):
             for token in generator:            
                 if token == model.token_eos() or stop_generation:
                     stop_generation = False
-                    app.logger.info('Abort generating')
+                    app.logger.info('End generating')
                     yield b''  # End of chunk
                     break
                     
@@ -143,7 +143,7 @@ def generate_search_request():
     tokens.append(LINEBREAK_TOKEN)        
     
     tokens = get_message_tokens(model=model, role="user", content=user_query[:200]) + [model.token_bos(), BOT_TOKEN, LINEBREAK_TOKEN]
-
+    stop_generation = False
     generator = model.generate(
         tokens,
         top_k=top_k,
@@ -215,6 +215,7 @@ def generate_response():
     app.logger.info('Prompt:')
     app.logger.info(model.detokenize(tokens[:CONTEXT_SIZE]).decode("utf-8", errors="ignore"))
 
+    stop_generation = False
     app.logger.info('Generate started')
     generator = model.generate(
         tokens[:CONTEXT_SIZE],