model_eval.py

import streamlit as st
import pandas as pd
import pickle


# Load the model
@st.cache_resource
def load_model():
    with open("data/model.pkl", "rb") as f:
        model = pickle.load(f)
        st.success("Model loaded successfully!")
        return model


# Page config
st.set_page_config(page_title="College Admission Predictor", page_icon="🎓")

st.title("College Admission Predictor")
st.write("Enter your information to see if you'll be admitted!")

# Load model
model = load_model()

# Create input form
with st.form("admission_form"):
    st.subheader("Your Academic Profile")

    col1, col2 = st.columns(2)

    with col1:
        gre = st.number_input("GRE Score", min_value=260, max_value=340, value=310)
        toefl = st.number_input("TOEFL Score", min_value=0, max_value=120, value=100)
        uni_rating = st.slider("University Rating", min_value=1, max_value=5, value=3)
        cgpa = st.number_input(
            "CGPA", min_value=0.0, max_value=10.0, value=8.0, step=0.1
        )

    with col2:
        sop = st.slider(
            "SOP Strength", min_value=1.0, max_value=5.0, value=3.0, step=0.5
        )
        lor = st.slider(
            "LOR Strength", min_value=1.0, max_value=5.0, value=3.0, step=0.5
        )
        research = st.selectbox(
            "Research Experience",
            options=[0, 1],
            format_func=lambda x: "Yes" if x == 1 else "No",
        )

    submitted = st.form_submit_button("Predict My Admission!")

# Make prediction when form is submitted
if submitted:
    # Create dataframe with user input
    user_data = pd.DataFrame(
        {
            "GRE Score": [gre],
            "TOEFL Score": [toefl],
            "University Rating": [uni_rating],
            "SOP": [sop],
            "LOR": [lor],
            "CGPA": [cgpa],
            "Research": [research],
        }
    )

    # Make prediction
    prediction = model.predict(user_data)[0]

    # Show result
    st.markdown("---")
    if prediction:
        st.success("**Congratulations! You're likely to be admitted!**")
        st.balloons()
    else:
        st.error("**Sorry, admission unlikely with current profile.**")
        st.write("Consider improving your scores or gaining research experience.")

    # Show the input data
    with st.expander("Your Profile Summary"):
        st.dataframe(user_data)