Added app.py

app.py

@@ -0,0 +1,47 @@
+import gradio as gr
+import torch
+from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
+
+# Set up device (GPU if available)
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+# Load the fine-tuned model and tokenizer
+model_name = "aarohanverma/text2sql-flan-t5-base-qlora-finetuned"  # Replace with your model repo name
+model = AutoModelForSeq2SeqLM.from_pretrained(model_name, torch_dtype=torch.bfloat16).to(device)
+tokenizer = AutoTokenizer.from_pretrained("google/flan-t5-base")
+
+def generate_sql(context: str, query: str) -> str:
+    """
+    Constructs a prompt using the user-provided context and query, then generates a SQL query.
+    """
+    prompt = f"""Context:
+{context}
+
+Query:
+{query}
+
+Response:
+"""
+    inputs = tokenizer(prompt, return_tensors="pt").to(device)
+    generated_ids = model.generate(
+        input_ids=inputs["input_ids"],
+        max_new_tokens=250,
+        temperature=0.0,      # Deterministic output
+        num_beams=3,          # Beam search for quality output
+        early_stopping=True,
+    )
+    return tokenizer.decode(generated_ids[0], skip_special_tokens=True)
+
+# Create a Gradio interface with two input boxes: one for context, one for query.
+iface = gr.Interface(
+    fn=generate_sql,
+    inputs=[
+        gr.Textbox(lines=8, label="Context", placeholder="Enter table schema, sample data, etc."),
+        gr.Textbox(lines=2, label="Query", placeholder="Enter your natural language query here...")
+    ],
+    outputs="text",
+    title="Text-to-SQL Generator",
+    description="Enter your own context (e.g., database schema and sample data) and a natural language query. The model will generate the corresponding SQL statement."
+)
+
+iface.launch()