Upload app.py

app.py

@@ -8,7 +8,7 @@ print(f"Using device: {device}")
 
 # Load the fine-tuned Phi-2 model in 4-bit quantization with CPU support
 #model_name = "./fine-tuned-phi2"
-model_name = "microsoft/phi-2"
+model_name = "microsoft/phi-2
 model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.float32, device_map={"": device})
 
 # Load tokenizer
@@ -23,25 +23,21 @@ def generate_response(prompt):
 
 # Define example prompts
 examples = [
-    ["What is machine learning?"],
+    ["What is the capital of France?"],
     ["Explain quantum mechanics in simple terms."],
     ["Write a short story about a robot discovering emotions."],
     ["Summarize the theory of relativity."]
 ]
 
 # Create Gradio UI
-# Launch Gradio app with a submit button
-# Create Gradio UI with a submit button
-with gr.Blocks() as app:
-    gr.Markdown("## Phi-2 LoRA Model")
-    gr.Markdown("A fine-tuned Phi-2 model with LoRA running on Hugging Face Spaces (CPU optimized).")
-    txt = gr.Textbox(lines=2, placeholder="Enter your prompt here...")
-    btn = gr.Button("Submit")
-    output = gr.Textbox()
-    
-    btn.click(fn=generate_response, inputs=txt, outputs=output)
-    
-    gr.Examples(examples=[[ex] for ex in examples], inputs=txt)
+iface = gr.Interface(
+    fn=generate_response,
+    inputs=gr.Textbox(lines=2, placeholder="Enter your prompt here..."),
+    outputs=gr.Textbox(),
+    title="Phi-2 LoRA Model",
+    description="A fine-tuned Phi-2 model with LoRA running on Hugging Face Spaces (CPU optimized).",
+    examples=examples,
+)
 
 # Launch Gradio app
-app.launch()
+iface.launch()