change model

app.py

@@ -55,9 +55,7 @@ def stream_transformer(message):
             early_stopping=True,
             seed=0,
         )
-        print("the original generate output:\n###\n")
-        print(tokenizer.decode(result[0], skip_special_tokens=True))
-        print("###\n")
+
         generator = model.generate(
             input_ids,
             max_new_tokens=20,
@@ -71,7 +69,7 @@ def stream_transformer(message):
             do_stream=True,
         )
         stream_result = ""
-        print("real-time stream chunk  generate output:\n###\n")
+
         words = ""
         last_tokens = []
         last_decoded_tokens = []
@@ -89,11 +87,6 @@ def stream_transformer(message):
                 last_decoded_tokens = tokens
 
             stream_result += word
-            print(f"chunk index: {index}: {word}")
-        print("###\n")
-        print("the stream cumulate generate output:\n###\n")
-        print(stream_result)
-        print("###\n")
-    
+        yield stream_result
     
  gr.ChatInterface(stream_transformer).launch()