Update app.py

app.py

@@ -1,12 +1,15 @@
 import gradio as gr
 from peft import AutoPeftModelForCausalLM
 from transformers import AutoTokenizer
+import spaces
 
-"""
-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
-"""
+# Load the model and tokenizer globally
+model = AutoPeftModelForCausalLM.from_pretrained("eforse01/lora_model", device_map="cpu")
+tokenizer = AutoTokenizer.from_pretrained("eforse01/lora_model")
 
-def respond(message, history: list[tuple[str, str]], system_message, max_tokens, temperature, min_p,):
+# Decorate the respond function with @spaces.GPU
+@spaces.GPU(duration=120)  # Set duration to 120 seconds if needed
+def respond(message, history: list[tuple[str, str]], system_message, max_tokens, temperature, min_p):
     messages = [{"role": "system", "content": system_message}]
 
     for val in history:
@@ -17,29 +20,27 @@ def respond(message, history: list[tuple[str, str]], system_message, max_tokens,
 
     messages.append({"role": "user", "content": message})
 
-    model = AutoPeftModelForCausalLM.from_pretrained(
-        "eforse01/lora_model", 
-        device_map="cpu",  
-    )
-
-    tokenizer = AutoTokenizer.from_pretrained("eforse01/lora_model")
     inputs = tokenizer.apply_chat_template(
         messages,
-        tokenize = True,
-        add_generation_prompt = True, 
-        return_tensors = "pt",
+        tokenize=True,
+        add_generation_prompt=True,
+        return_tensors="pt",
     )
 
-    output = model.generate(input_ids = inputs, max_new_tokens = max_tokens,
-                    use_cache = True, temperature = temperature, min_p = min_p)
+    # Generate response
+    output = model.generate(
+        input_ids=inputs.input_ids,
+        max_new_tokens=max_tokens,
+        use_cache=True,
+        temperature=temperature,
+        min_p=min_p,
+    )
     
+    # Decode and format response
     response = tokenizer.batch_decode(output, skip_special_tokens=True)[0]
-
     yield response.split('assistant')[-1]
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
+# Gradio Interface
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[
@@ -57,4 +58,4 @@ demo = gr.ChatInterface(
 )
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()