app.py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
from datasets import load_dataset
import random

# Load the Spider dataset
spider_dataset = load_dataset("spider", split='train')

# Load tokenizer and model
tokenizer = AutoTokenizer.from_pretrained("mrm8488/t5-base-finetuned-wikiSQL")
model = AutoModelForSeq2SeqLM.from_pretrained("mrm8488/t5-base-finetuned-wikiSQL")

def generate_sql(query):
    input_text = "translate English to SQL: " + query
    inputs = tokenizer(input_text, return_tensors="pt", padding=True)
    outputs = model.generate(**inputs, max_length=512)
    sql_query = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return sql_query

def get_random_example():
    example = random.choice(spider_dataset)
    return example['question'], generate_sql(example['question'])

# Create a Gradio interface
interface = gr.Interface(
    fn=generate_sql,
    inputs=gr.Textbox(lines=2, placeholder="Enter your natural language query here..."),
    outputs="text",
    title="NL to SQL with T5",
    description="This model converts natural language queries into SQL. Enter your query or get a random example from the Spider dataset!"
)

# Add a button to get a random example from the dataset
interface.add_component(
    gr.Button("Get Random Example"),
    fn=get_random_example,
    inputs=[],
    outputs=[gr.Textbox(label="Random Question"), gr.Textbox(label="Generated SQL")]
)

# Launch the app
if __name__ == "__main__":
    interface.launch()