Update app.py

app.py

@@ -2,17 +2,24 @@ import gradio as gr
 import torch
 from peft import PeftModel, PeftConfig
 
+
+peft_model_id = f"IThinkUPC/SQLGenerator-AI"
+config = PeftConfig.from_pretrained(peft_model_id)
+model = AutoModelForCausalLM.from_pretrained(config.base_model_name_or_path, return_dict=True, load_in_8bit=True, device_map='auto')
+tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path)
+# Load the Lora model
+model = PeftModel.from_pretrained(model, peft_model_id)
+
 def greet(name):
     return "Hello " + name + "!!"
 
-def model_run():
-    peft_model_id = f"IThinkUPC/SQLGenerator-AI"
-    config = PeftConfig.from_pretrained(peft_model_id)
-    model = AutoModelForCausalLM.from_pretrained(config.base_model_name_or_path, return_dict=True, load_in_8bit=True, device_map='auto')
-    tokenizer = AutoTokenizer.from_pretrained(config.base_model_name_or_path)
-    # Load the Lora model
-    model = PeftModel.from_pretrained(model, peft_model_id)
+def make_inference(prompt):  
+  batch = tokenizer(f"### Question:\n{prompt}: \n\n### Query", return_tensors='pt')
+  with torch.cuda.amp.autocast():
+    output_tokens = model.generate(**batch, max_new_tokens=50)
+  dreturn tokenizer.decode(output_tokens[0], skip_special_tokens=True)
     
 
-iface = gr.Interface(fn=greet, inputs="text", outputs="text")
+#iface = gr.Interface(fn=greet, inputs="text", outputs="text")
+iface = gr.Interface(fn=make_inference, inputs="text", outputs="text")
 iface.launch()