Update app.py

app.py

@@ -1,10 +1,9 @@
 import pandas as pd
 import numpy as np
 import gradio as gr
-pip install streamlit
-
-import streamlit as st
+import torch
 from transformers import AutoModelForMultipleChoice, AutoTokenizer
+from huggingface_hub import hf_hub_url, Repository
 
 # Load the model and tokenizer
 model_path = "/kaggle/input/deberta-v3-large-hf-weights"
@@ -43,32 +42,10 @@ iface = gr.Interface(
     description="Enter the prompt and options (A to E) below and get predictions.",
 )
 
-# Function to run Gradio Interface using Streamlit
-def run_gradio_interface():
-    st.title("LLM Science Exam Demo")
-    st.markdown("Enter the prompt and options (A to E) below and get predictions.")
-    
-    prompt = st.text_area("Prompt", value="This is the prompt", height=100)
-    option_a = st.text_input("Option A", value="Option A text")
-    option_b = st.text_input("Option B", value="Option B text")
-    option_c = st.text_input("Option C", value="Option C text")
-    option_d = st.text_input("Option D", value="Option D text")
-    option_e = st.text_input("Option E", value="Option E text")
-    
-    sample_data = {
-        "prompt": prompt,
-        "A": option_a,
-        "B": option_b,
-        "C": option_c,
-        "D": option_d,
-        "E": option_e,
-    }
-    
-    predictions = iface.process(sample_data)
-    st.markdown("### Predictions:")
-    st.write(predictions)
+# Run the interface locally
+iface.launch(share=True)
 
-# Run the Streamlit app
-if __name__ == "__main__":
-    st.set_page_config(page_title="LLM Science Exam Demo", page_icon=":memo:")
-    run_gradio_interface()
+# Once you have verified that the interface works as expected, proceed to create the Hugging Face space:
+repo_url = hf_hub_url("your-username/your-repo-name")
+repo = Repository.from_hf_hub(repo_url)
+repo.push(path="./my_model", model=model, tokenizer=tokenizer, config=model.config)