Update app.py

app.py

@@ -267,28 +267,28 @@ def simple_chat(message: dict, temperature: float = 0.8, max_length: int = 4096,
             eos_token_id=[151329, 151336, 151338],
         )
 
-        gen_kwargs = {**input_ids, **generate_kwargs}
-
-        with torch.no_grad():
-            thread = Thread(target=model.generate, kwargs=gen_kwargs)
-            thread.start()
-            buffer = ""
-            for new_text in streamer:
-                buffer += new_text
-                yield buffer
-
-        print("--------------")
-        print("Buffer: ")
-        print("  ")
-        print(buffer)
-        print("  ")
-        print("--------------")
+        # gen_kwargs = {**input_ids, **generate_kwargs}
 
         # with torch.no_grad():
-        #     generated_ids = model.generate(input_ids['input_ids'], **generate_kwargs)
-        #     generated_text = tokenizer.decode(generated_ids[0], skip_special_tokens=True)
+        #     thread = Thread(target=model.generate, kwargs=gen_kwargs)
+        #     thread.start()
+        #     buffer = ""
+        #     for new_text in streamer:
+        #         buffer += new_text
+        #         yield buffer
+
+        # print("--------------")
+        # print("Buffer: ")
+        # print("  ")
+        # print(buffer)
+        # print("  ")
+        # print("--------------")
+
+        with torch.no_grad():
+            generated_ids = model.generate(input_ids['input_ids'], **generate_kwargs)
+            generated_text = tokenizer.decode(generated_ids[0], skip_special_tokens=True)
 
-        generated_text = buffer
+        #generated_text = buffer
 
 
         return PlainTextResponse(generated_text)