Update app.py

app.py

@@ -1,11 +1,9 @@
 import gradio as gr
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
-
-#change the directory address below based on what we call the output QLORA model
-
-tokenizer = AutoTokenizer.from_pretrained("<lehmanc25>/my-phi2-model")
-model = AutoModelForCausalLM.from_pretrained("<lehmanc25>/my-phi2-model")
+# Assuming you've uploaded your model to Hugging Face with the username 'lehmanc25'
+tokenizer = AutoTokenizer.from_pretrained("lehmanc25/my-phi2-model")
+model = AutoModelForCausalLM.from_pretrained("lehmanc25/my-phi2-model")
 
 def chat_with_phi2(input_text):
     encoded_input = tokenizer(input_text, return_tensors='pt')
@@ -13,14 +11,19 @@ def chat_with_phi2(input_text):
     response = tokenizer.decode(output[0], skip_special_tokens=True)
     return response
 
-interface = gr.Interface(
-    fn=chat_with_phi2,
-    inputs=gr.Textbox(lines=2, placeholder="Ask something about W&L..."),
-    
-    #model(inputs)
-    outputs="text",
-    title="W&L ChatGPT-like AI",
-    description="Ask any question about Washington and Lee University!"
-)
+# Using Gradio Blocks to build a more dynamic interface
+block = gr.Blocks()
 
+with block:
+    gr.Markdown("# W&L ChatGPT-like AI")
+    gr.Markdown("Ask any question about Washington and Lee University!")
+    input_text = gr.Textbox(label="Your question:", lines=2, placeholder="Ask something about W&L...")
+    output_text = gr.Text(label="AI Response:")
+    button = gr.Button("Ask")
+    button.click(
+        fn=chat_with_phi2, 
+        inputs=input_text, 
+        outputs=output_text
+    )
 
+block.launch()