Update app.py

app.py

@@ -30,7 +30,46 @@ def optimize(code, category):
     if not code.strip():
         return "Please enter some Python code."
     
+    # Using triple quotes f""" here prevents the "unterminated string" error
+    user_content = f"""Original Code:
+```python
+{code}
+```
+Category: {category}"""
+
     messages = [
         {"role": "system", "content": SYSTEM_PROMPT},
-        {"role": "user", "content": f"Original Code:\n
-http://googleusercontent.com/immersive_entry_chip/0
+        {"role": "user", "content": user_content}
+    ]
+    
+    text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
+    model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
+    
+    with torch.no_grad():
+        generated_ids = model.generate(
+            **model_inputs, 
+            max_new_tokens=1024,
+            do_sample=True,
+            temperature=0.7,
+            top_p=0.95,
+            pad_token_id=tokenizer.eos_token_id
+        )
+    
+    input_len = model_inputs["input_ids"].shape[1]
+    output_ids = generated_ids[0][input_len:]
+    return tokenizer.decode(output_ids, skip_special_tokens=True)
+
+demo = gr.Interface(
+    fn=optimize,
+    inputs=[
+        gr.Code(language="python", label="Your Python Code", lines=15),
+        gr.Radio(choices=["Performance", "Readability", "Conciseness"], value="Performance", label="Optimization Category")
+    ],
+    outputs=gr.Textbox(label="Optimized Code & Explanation", lines=20),
+    title="⚡ Python Code Optimizer",
+    description="QLoRA fine-tuned Qwen2.5-Coder-7B.",
+    flagging_mode="never"
+)
+
+if __name__ == "__main__":
+    demo.launch()