changes in message encoding

app.py

@@ -22,13 +22,14 @@ def predict(message, history):
     history_transformer_format = history + [[message, ""]]
     stop = StopOnTokens()
 
-    messages = "".join(["".join(["<|start_header_id|>user<|end_header_id|>\n"+item[0], 
-                                 "<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n"+item[1]])
-                for item in history_transformer_format])
-    print("MESSAGES: ", messages)
-
-    model_inputs = tokenizer([messages], return_tensors="pt") # .to("cuda")
-    print("MODEL INPUT:\n", model_inputs)
+    # messages = "".join(["".join(["<|start_header_id|>user<|end_header_id|>\n"+item[0], 
+    #                              "<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n"+item[1]])
+    #             for item in history_transformer_format])
+    messages = "".join(["".join(["\n<human>:"+item[0], "\n<bot>:"+item[1]])
+                        for item in history_transformer_format])
+
+    # model_inputs = tokenizer([messages], return_tensors="pt") # .to("cuda")
+    model_inputs = tokenizer.apply_chat_template(messages, return_tensors="pt")
     streamer = TextIteratorStreamer(tokenizer, timeout=10., skip_prompt=True, skip_special_tokens=True)
     generate_kwargs = dict(
         model_inputs,