Update app.py

app.py

@@ -1,11 +1,13 @@
 import gradio as gr
-from transformers import AutoModelForCausalLM, AutoTokenizer
-import torch
+from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
+import onnxruntime as ort
 
 # Load the Phi-3.5-mini-instruct model and tokenizer
-model_name = "phi-3.5-mini-instruct"
+model_name = "microsoft/Phi-3.5-mini-instruct"
 tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForCausalLM.from_pretrained(model_name)
+
+# Load the ONNX model
+session = ort.InferenceSession(f"{model_name}/model.onnx")
 
 # Simple HTML template for the website
 simple_website_template = """
@@ -42,10 +44,15 @@ def personalize_website_llm(persona_text):
     # Create a prompt for the model
     prompt = f"Generate personalized website content for the following persona: {persona_text}. Provide a title and main content."
     
-    # Tokenize and generate output
-    inputs = tokenizer(prompt, return_tensors="pt")
-    outputs = model.generate(inputs.input_ids, max_length=150)
-    generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    # Tokenize the prompt
+    inputs = tokenizer(prompt, return_tensors="np")
+
+    # Run the ONNX model
+    ort_inputs = {session.get_inputs()[0].name: inputs["input_ids"]}
+    ort_outs = session.run(None, ort_inputs)
+    
+    # Decode the output
+    generated_text = tokenizer.decode(ort_outs[0][0], skip_special_tokens=True)
     
     # Split the response into a title and content
     title, content = generated_text.split('\n', 1)