app.py

import gradio as gr
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

# Load the fine-tuned model and tokenizer
def load_model_and_tokenizer(model_dir):
    try:
        tokenizer = AutoTokenizer.from_pretrained(model_dir)
        model = AutoModelForCausalLM.from_pretrained(model_dir, torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32)
        if torch.cuda.is_available():
            model = model.to('cuda')
        else:
            model = model.to('cpu')
    except Exception as e:
        print(f"Error loading model: {e}")
        return None, None
    return model, tokenizer

# Update the model directory path to use forward slashes
model_dir = "KJX123/Llama2-7b-finetune"
model, tokenizer = load_model_and_tokenizer(model_dir)

# Function to generate code
def generate_code(query):
    if not model or not tokenizer:
        return "Model or tokenizer not loaded properly."
    
    prompt = f"Query: {query}\nGitHub Code:\nYouTube Code:"
    inputs = tokenizer(prompt, return_tensors='pt')
    
    device = 'cuda' if torch.cuda.is_available() else 'cpu'
    inputs = inputs.to(device)
    
    outputs = model.generate(inputs['input_ids'], max_length=600, num_return_sequences=1)
    generated_code = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return generated_code

def gradio_interface(query):
    code = generate_code(query)
    return code

# Define Gradio app layout
iface = gr.Interface(
    fn=gradio_interface,
    inputs=gr.Textbox(lines=2, placeholder="Enter your query here..."),
    outputs=gr.Textbox(lines=20, placeholder="Generated code will appear here..."),
    title="Code Generator",
    description="Enter a programming task or query to generate code using the fine-tuned model."
)

if __name__ == "__main__":
    iface.launch()