textstreamer

app.py

@@ -1,6 +1,6 @@
 from peft import AutoPeftModelForCausalLM
-from transformers import AutoTokenizer
-import gradio as gr
+from transformers import AutoTokenizer, TextStreamer
+# import gradio as gr
 """
 For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
 """
@@ -12,19 +12,19 @@ model = AutoPeftModelForCausalLM.from_pretrained(
     )
 tokenizer = AutoTokenizer.from_pretrained("EITD/lora_model_1")
 
-# messages = [{"role": "user", "content": "Continue the Fibonacci sequence: 1, 1, 2, 3, 5, 8,"},]
+messages = [{"role": "user", "content": "Continue the Fibonacci sequence: 1, 1, 2, 3, 5, 8,"},]
 
-# inputs = tokenizer.apply_chat_template(
-#     messages,
-#     tokenize = True,
-#     add_generation_prompt = True, # Must add for generation
-#     return_tensors = "pt",
-# )
+inputs = tokenizer.apply_chat_template(
+    messages,
+    tokenize = True,
+    add_generation_prompt = True, # Must add for generation
+    return_tensors = "pt",
+)
 
-# outputs = model.generate(input_ids = inputs, max_new_tokens = 64, use_cache = True,
-#                         temperature = 1.5, min_p = 0.1)
+outputs = model.generate(input_ids = inputs, max_new_tokens = 64, use_cache = True,
+                        temperature = 1.5, min_p = 0.1)
 
-# print(tokenizer.batch_decode(outputs))
+print(tokenizer.batch_decode(outputs))
 
 def respond(
     message,
@@ -65,30 +65,35 @@ def respond(
         return_tensors = "pt",
     )
     
-    outputs = model.generate(input_ids = inputs, max_new_tokens = max_tokens, use_cache = True,
-                         temperature = temperature, min_p = top_p)
-    return tokenizer.batch_decode(outputs)
+    # outputs = model.generate(input_ids = inputs, max_new_tokens = max_tokens, use_cache = True,
+    #                      temperature = temperature, min_p = top_p)
+    
+    text_streamer = TextStreamer(tokenizer, skip_prompt = True)
+    model.generate(input_ids = inputs, streamer = text_streamer, max_new_tokens = max_tokens,
+                    use_cache = True, temperature = temperature, min_p = top_p)
+    
+    # return tokenizer.batch_decode(outputs)
 
 
 """
 For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
 """
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
+# demo = gr.ChatInterface(
+#     respond,
+#     additional_inputs=[
+#         gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
+#         gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
+#         gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
+#         gr.Slider(
+#             minimum=0.1,
+#             maximum=1.0,
+#             value=0.95,
+#             step=0.05,
+#             label="Top-p (nucleus sampling)",
+#         ),
+#     ],
+# )
 
 
-if __name__ == "__main__":
-    demo.launch()
+# if __name__ == "__main__":
+#     demo.launch()