Spaces:
Running
Running
nileshhanotia
commited on
Commit
•
cacc96f
1
Parent(s):
0d53dda
Update app.py
Browse files
app.py
CHANGED
@@ -1,13 +1,72 @@
|
|
1 |
import gradio as gr
|
|
|
|
|
2 |
from transformers import AutoModelForCausalLM, AutoTokenizer
|
3 |
|
|
|
4 |
model_name = "premai-io/prem-1B-SQL"
|
5 |
-
tokenizer = AutoTokenizer.from_pretrained(
|
6 |
-
model = AutoModelForCausalLM.from_pretrained(
|
7 |
|
8 |
def generate_sql(natural_language_query):
|
9 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
10 |
return sql_query
|
11 |
|
12 |
-
|
13 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
import gradio as gr
|
2 |
+
import mysql.connector
|
3 |
+
from mysql.connector import Error
|
4 |
from transformers import AutoModelForCausalLM, AutoTokenizer
|
5 |
|
6 |
+
# Load the model and tokenizer
|
7 |
model_name = "premai-io/prem-1B-SQL"
|
8 |
+
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
9 |
+
model = AutoModelForCausalLM.from_pretrained(model_name)
|
10 |
|
11 |
def generate_sql(natural_language_query):
|
12 |
+
"""Generate SQL query from natural language."""
|
13 |
+
# Define your schema information
|
14 |
+
schema_info = """
|
15 |
+
CREATE TABLE sales (
|
16 |
+
pizza_id DECIMAL(8,2) PRIMARY KEY,
|
17 |
+
order_id DECIMAL(8,2),
|
18 |
+
pizza_name_id VARCHAR(14),
|
19 |
+
quantity DECIMAL(4,2),
|
20 |
+
order_date DATE,
|
21 |
+
order_time VARCHAR(8),
|
22 |
+
unit_price DECIMAL(5,2),
|
23 |
+
total_price DECIMAL(5,2),
|
24 |
+
pizza_size VARCHAR(3),
|
25 |
+
pizza_category VARCHAR(7),
|
26 |
+
pizza_ingredients VARCHAR(97),
|
27 |
+
pizza_name VARCHAR(42)
|
28 |
+
);
|
29 |
+
"""
|
30 |
+
|
31 |
+
# Construct the prompt
|
32 |
+
prompt = f"""### Task: Generate a SQL query to answer the following question.
|
33 |
+
|
34 |
+
### Database Schema:
|
35 |
+
{schema_info}
|
36 |
+
|
37 |
+
### Question: {natural_language_query}
|
38 |
+
|
39 |
+
### SQL Query:"""
|
40 |
+
|
41 |
+
# Tokenize and generate
|
42 |
+
inputs = tokenizer(prompt, return_tensors="pt", add_special_tokens=False).to(model.device)
|
43 |
+
outputs = model.generate(
|
44 |
+
inputs["input_ids"],
|
45 |
+
max_length=512,
|
46 |
+
temperature=0.1,
|
47 |
+
do_sample=True,
|
48 |
+
top_p=0.95,
|
49 |
+
num_return_sequences=1,
|
50 |
+
eos_token_id=tokenizer.eos_token_id,
|
51 |
+
pad_token_id=tokenizer.pad_token_id
|
52 |
+
)
|
53 |
+
|
54 |
+
# Decode and clean up the response
|
55 |
+
generated_query = tokenizer.decode(outputs[0], skip_special_tokens=True)
|
56 |
+
sql_query = generated_query.split("### SQL Query:")[-1].strip()
|
57 |
+
|
58 |
return sql_query
|
59 |
|
60 |
+
def main():
|
61 |
+
# Gradio interface setup
|
62 |
+
iface = gr.Interface(
|
63 |
+
fn=generate_sql,
|
64 |
+
inputs="text",
|
65 |
+
outputs="text",
|
66 |
+
title="Natural Language to SQL Query Generator",
|
67 |
+
description="Enter a natural language query to generate the corresponding SQL query."
|
68 |
+
)
|
69 |
+
iface.launch()
|
70 |
+
|
71 |
+
if __name__ == "__main__":
|
72 |
+
main()
|