Spaces:
Sleeping
Sleeping
Create app.py
Browse files
app.py
ADDED
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import gradio as gr
|
2 |
+
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
|
3 |
+
from datasets import load_dataset
|
4 |
+
|
5 |
+
# Load the Spider dataset
|
6 |
+
spider_dataset = load_dataset("HusnaManakkot/new-spider-HM", split='train') # Load a subset of the dataset
|
7 |
+
# Extract schema information from the Spider dataset
|
8 |
+
table_names = set()
|
9 |
+
column_names = set()
|
10 |
+
for item in spider_dataset:
|
11 |
+
for table in item['db_id']:
|
12 |
+
table_names.add(table)
|
13 |
+
for column in item['question']:
|
14 |
+
column_names.add(column)
|
15 |
+
|
16 |
+
# Load tokenizer and model
|
17 |
+
tokenizer = AutoTokenizer.from_pretrained("mrm8488/t5-base-finetuned-wikiSQL") # Update this to a model fine-tuned on Spider if available
|
18 |
+
model = AutoModelForSeq2SeqLM.from_pretrained("mrm8488/t5-base-finetuned-wikiSQL") # Update this to a model fine-tuned on Spider if available
|
19 |
+
|
20 |
+
def generate_sql_from_user_input(query):
|
21 |
+
# Generate SQL for the user's query
|
22 |
+
input_text = "translate English to SQL: " + query
|
23 |
+
inputs = tokenizer(input_text, return_tensors="pt", padding=True)
|
24 |
+
outputs = model.generate(**inputs, max_length=512)
|
25 |
+
sql_query = tokenizer.decode(outputs[0], skip_special_tokens=True)
|
26 |
+
|
27 |
+
# Post-process the SQL query to match the dataset's schema
|
28 |
+
for table_name in table_names:
|
29 |
+
if "TABLE" in sql_query:
|
30 |
+
sql_query = sql_query.replace("TABLE", table_name)
|
31 |
+
break # Assuming only one table is referenced in the query
|
32 |
+
for column_name in column_names:
|
33 |
+
if "COLUMN" in sql_query:
|
34 |
+
sql_query = sql_query.replace("COLUMN", column_name, 1)
|
35 |
+
return sql_query
|
36 |
+
|
37 |
+
# Create a Gradio interface
|
38 |
+
interface = gr.Interface(
|
39 |
+
fn=generate_sql_from_user_input,
|
40 |
+
inputs=gr.Textbox(label="Enter your natural language query"),
|
41 |
+
outputs=gr.Textbox(label="Generated SQL Query"),
|
42 |
+
title="NL to SQL with T5 using Spider Dataset",
|
43 |
+
description="This model generates an SQL query for your natural language input based on the Spider dataset."
|
44 |
+
)
|
45 |
+
|
46 |
+
# Launch the app
|
47 |
+
if __name__ == "__main__":
|
48 |
+
interface.launch()
|