Spaces:
Sleeping
Sleeping
File size: 1,480 Bytes
35849b2 5822624 bb4a570 47a2dc5 bb4a570 ee62733 bb4a570 ee62733 35849b2 55d6351 35849b2 94ca4fb 55d6351 5822624 ee62733 47a2dc5 35849b2 bb4a570 ee62733 5822624 ee62733 35849b2 5822624 ee62733 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 |
# app.py
import gradio as gr
from transformers import T5Tokenizer, T5ForConditionalGeneration
from deep_translator import GoogleTranslator
import torch
torch.set_num_threads(1)
torch.set_num_interop_threads(1)
model_name = "mrm8488/t5-base-finetuned-wikiSQL"
tokenizer = T5Tokenizer.from_pretrained(model_name)
model = T5ForConditionalGeneration.from_pretrained(model_name)
# Tu esquema personalizado
SCHEMA = """
Schema:
Table bodegas with columns: Id, Nombre, Encargado, Telefono, Email, Direccion, Horario, Regional, Latitud, Longitud.
Table maestra with columns: CodigoSap, Descripcion, Grupo, Agrupador, Marca, Parte, Operacion, Componente.
"""
def generar_sql(pregunta_espanol):
try:
pregunta_ingles = GoogleTranslator(source="es", target="en").translate(pregunta_espanol)
prompt = f"{SCHEMA}\ntranslate English to SQL: {pregunta_ingles}"
input_ids = tokenizer(prompt, return_tensors="pt").input_ids
output = model.generate(input_ids, max_length=128)
sql = tokenizer.decode(output[0], skip_special_tokens=True)
return sql
except Exception as e:
return f"Error: {str(e)}"
iface = gr.Interface(
fn=generar_sql,
inputs=gr.Textbox(lines=3, label="Pregunta en español"),
outputs=gr.Textbox(label="Consulta SQL generada"),
title="Texto a SQL con esquema personalizado",
description="Escribe una pregunta en español y genera SQL sobre las tablas `bodegas` y `maestra`."
)
iface.launch()
|