import streamlit as st import numpy as np import joblib # ----------------------------- # Page config # ----------------------------- st.set_page_config( page_title="Student Grade Prediction", page_icon="🎓", layout="centered" ) # ----------------------------- # Load model # ----------------------------- @st.cache_resource def load_model(): model = joblib.load("src/gb_model.pkl") return model model = load_model() # ----------------------------- # Sidebar # ----------------------------- st.sidebar.title("About this app") st.sidebar.markdown( """ This app uses a **Gradient Boosting Regressor** to predict a student's **marks**. **Inputs:** - Number of courses - Study time per day (hours) **Output:** - Predicted marks """ ) st.sidebar.markdown("---") st.sidebar.markdown("Made for a machine learning regression project. 🎓") # ----------------------------- # Main title # ----------------------------- st.title("🎓 Student Grade Prediction App") st.markdown( """ This app predicts the **marks of a student** based on the **number of courses** and the **study time per day**. The model was trained using **Gradient Boosting Regression**. """ ) # ----------------------------- # Input area # ----------------------------- st.subheader("Enter student information") col1, col2 = st.columns(2) with col1: number_courses = st.number_input( "Number of courses", min_value=1, max_value=20, value=5, step=1, help="How many courses the student is taking." ) with col2: time_study = st.number_input( "Study time per day (hours)", min_value=0.0, max_value=12.0, value=4.0, step=0.5, help="Average study time per day in hours." ) # ----------------------------- # Prediction # ----------------------------- st.markdown("---") if st.button("🔮 Predict Marks"): # create feature array for model features = np.array([[number_courses, time_study]]) # prediction predicted_marks = model.predict(features)[0] st.success(f"Predicted marks: **{predicted_marks:.2f}**") st.markdown( """ **Interpretation (simple):** If the student keeps this study time and number of courses, the model expects marks close to this value. """ ) # ----------------------------- # Extra info # ----------------------------- with st.expander("Show model details (simple explanation)"): st.markdown( """ - Model type: **GradientBoostingRegressor** - Target: **Marks** - Main feature: **Study time per day** - Secondary feature: **Number of courses** """ )