app.py

import gradio as gr


def generate_sql(schema, query):
    messages = [
        {"role": "system", "content": f"""You are a SQL assistant. Use the following database schema to answer the user's questions:\n{schema}"""},
        {"role": "user", "content": query},
    ]
    inputs = tokenizer.apply_chat_template(
        messages,
        tokenize=True,
        add_generation_prompt=True,
        return_tensors="pt",
    ).to("cuda")

    text_streamer = TextStreamer(tokenizer, skip_prompt=True)
    outputs = model.generate(
        input_ids=inputs,
        streamer=text_streamer,
        max_new_tokens=400,
        use_cache=True,
        temperature=0.1,  # Adjust temperature as needed
        min_p=0.1,
    )
    generated_text = tokenizer.batch_decode(outputs)[0]
    return generated_text


iface = gr.Interface(
    fn=generate_sql,
    inputs=[
        gr.Textbox(label="Database Schema", lines=5),
        gr.Textbox(label="SQL Query"),
    ],
    outputs=gr.Textbox(label="Generated SQL"),
    title="SQL Assistant",
    description="Enter a database schema and a SQL query to get the corresponding SQL code.",
)

iface.launch()