Update app.py

app.py

@@ -1,68 +1,64 @@
 import os
 os.system('sh setup.sh')
 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
-"""
-import os
-os.system('sh setup.sh')
-
-client = InferenceClient("redael/model_udc")
+import torch
+from transformers import GPT2LMHeadModel, GPT2Tokenizer
 
+# Load the model and tokenizer
+model_path = "final_model"
+tokenizer = GPT2Tokenizer.from_pretrained(model_path)
+model = GPT2LMHeadModel.from_pretrained(model_path)
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+# Check if CUDA is available and use GPU if possible
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+model.to(device)
 
-    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,
+def generate_response(prompt, model, tokenizer, max_length=100, num_beams=5, temperature=0.5, top_p=0.9, repetition_penalty=4.0):
+    # Prepare the prompt
+    prompt = f"User: {prompt}\nAssistant:"
+    inputs = tokenizer(prompt, return_tensors='pt', padding=True, truncation=True, max_length=512).to(device)
+    outputs = model.generate(
+        inputs['input_ids'],
+        max_length=max_length,
+        num_return_sequences=1,
+        pad_token_id=tokenizer.eos_token_id,
+        num_beams=num_beams,
         temperature=temperature,
         top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-
-        response += token
-        yield response
+        repetition_penalty=repetition_penalty,
+        early_stopping=True
+    )
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    
+    # Post-processing to clean up the response
+    response = response.split("Assistant:")[-1].strip()
+    response_lines = response.split('\n')
+    clean_response = []
+    for line in response_lines:
+        if "User:" not in line and "Assistant:" not in line:
+            clean_response.append(line)
+    response = ' '.join(clean_response)
+    return response.strip()
 
-"""
-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)",
-        ),
-    ],
-)
+def chat_interface(user_input, history):
+    response = generate_response(user_input, model, tokenizer)
+    history.append((user_input, response))
+    return history, history
 
+with gr.Blocks() as demo:
+    gr.Markdown("# Chatbot using GPT")
+    
+    chatbot = gr.Chatbot()
+    message = gr.Textbox(placeholder="Type your question here...")
+    state = gr.State([])
+    
+    with gr.Row():
+        clear = gr.Button("Clear")
+        submit = gr.Button("Send")
+    
+    submit.click(chat_interface, [message, state], [chatbot, state])
+    clear.click(lambda: None, None, chatbot)
+    clear.click(lambda: [], None, state)
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()