Update app.py

app.py

@@ -1,11 +1,7 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
-
-"""
-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
-"""
-client = InferenceClient("UpMath/Thai-HomeworkGen-v3")
+from transformers import pipeline
 
+pipe = pipeline(task="text-generation", model="UpMath/Thai-HomeworkGen-v3")
 
 def respond(
     message,
@@ -15,34 +11,27 @@ def respond(
     temperature,
     top_p,
 ):
-    messages = [{"role": "system", "content": system_message}]
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
+    context = system_message + "\n"
+    for user, assistant in history:
+        if user:
+            context += f"User: {user}\n"
+        if assistant:
+            context += f"Assistant: {assistant}\n"
+    context += f"User: {message}\nAssistant:"
+
+    outputs = pipe(
+        context,
+        max_new_tokens=max_tokens,
         temperature=temperature,
         top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-
-        response += token
-        yield response
+        do_sample=True,
+    )
 
+    response = outputs[0]["generated_text"]
+    
+    response_only = response[len(context):].strip()
+    yield response_only
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[
@@ -59,6 +48,5 @@ demo = gr.ChatInterface(
     ],
 )
 
-
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()