Upload inference.py with huggingface_hub

inference.py

@@ -0,0 +1,22 @@
+
+# Example inference code
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+# Load the model
+tokenizer = AutoTokenizer.from_pretrained("exaler/aaa-2-sql-2")
+model = AutoModelForCausalLM.from_pretrained("exaler/aaa-2-sql-2")
+
+def generate_sql(instruction, input_text):
+    # Format prompt
+    prompt = f"<s>[INST] {instruction}\n\n{input_text} [/INST]"
+    
+    # Generate
+    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+    outputs = model.generate(
+        inputs=inputs.input_ids, 
+        max_new_tokens=512,
+        temperature=0.0,
+        do_sample=False
+    )
+    response = tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True)
+    return response