Using custom model

app.py

@@ -1,52 +1,48 @@
 import gradio as gr
-from transformers import AutoModelForCausalLM, AutoTokenizer
-from typing import List, Tuple
+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("DominusDeorum/llama-3.2-lora_model")
 
-# Load your fine-tuned model and tokenizer
-model_name = "DominusDeorum/llama-3.2-lora_model"  # Replace with your model name
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForCausalLM.from_pretrained(model_name)
 
-# The main function that handles the chatbot response
 def respond(
-    message: str,
-    history: List[Tuple[str, str]],
-    system_message: str,
-    max_tokens: int,
-    temperature: float,
-    top_p: float,
+    message,
+    history: list[tuple[str, str]],
+    system_message,
+    max_tokens,
+    temperature,
+    top_p,
 ):
-    # Start with the system message
     messages = [{"role": "system", "content": system_message}]
 
-    # Add the user-assistant history to the messages
     for val in history:
         if val[0]:
             messages.append({"role": "user", "content": val[0]})
         if val[1]:
             messages.append({"role": "assistant", "content": val[1]})
 
-    # Add the new user message
     messages.append({"role": "user", "content": message})
 
-    # Tokenize the conversation history for model input
-    inputs = tokenizer(" ".join([msg['content'] for msg in messages]), return_tensors="pt")
+    response = ""
 
-    # Generate the response using the fine-tuned model
-    outputs = model.generate(
-        inputs.input_ids, 
-        max_length=max_tokens, 
-        temperature=temperature, 
-        top_p=top_p
-    )
+    for message in client.chat_completion(
+        messages,
+        max_tokens=max_tokens,
+        stream=True,
+        temperature=temperature,
+        top_p=top_p,
+    ):
+        token = message.choices[0].delta.content
 
-    # Decode the generated response tokens
-    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+        response += token
+        yield response
 
-    # Yield the final response to the user
-    return response
 
-# Set up the Gradio interface for the chatbot
+"""
+For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
+"""
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[
@@ -63,6 +59,6 @@ demo = gr.ChatInterface(
     ],
 )
 
-# Launch the interface
+
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()