testing max_length

app.py

@@ -58,7 +58,7 @@ def gpt_respond(tab_name, message, chat_history, max_convo_length = 10):
 def vicuna_respond(tab_name, message, chat_history):
     formatted_prompt = f'''Generate the output only for the assistant. Please output any <{tab_name}> in the following sentence one per line without any additional text: <{message}>'''
     input_ids = vicuna_tokenizer.encode(formatted_prompt, return_tensors="pt")
-    output_ids = vicuna_model.generate(input_ids, num_beams=5, no_repeat_ngram_size=2)
+    output_ids = vicuna_model.generate(input_ids, max_length=149, num_beams=5, no_repeat_ngram_size=2)
     bot_message = vicuna_tokenizer.decode(output_ids[0], skip_special_tokens=True)
         
     chat_history.append((formatted_prompt, bot_message))
@@ -68,7 +68,7 @@ def vicuna_respond(tab_name, message, chat_history):
 def llama_respond(tab_name, message, chat_history):
     formatted_prompt = f'''Generate the output only for the assistant. Please output any <{tab_name}> in the following sentence one per line without any additional text: <{message}>'''
     input_ids = llama_tokenizer.encode(formatted_prompt, return_tensors="pt")
-    output_ids = llama_model.generate(input_ids, num_beams=5, no_repeat_ngram_size=2)
+    output_ids = llama_model.generate(input_ids, max_length=149, num_beams=5, no_repeat_ngram_size=2)
     bot_message = llama_tokenizer.decode(output_ids[0], skip_special_tokens=True)
         
     chat_history.append((formatted_prompt, bot_message))