import numpy as np import pandas as pd import streamlit as st import joblib from pathlib import Path st.set_page_config( page_title='Flood Probability Predictor (Linear Regression)', page_icon='🌊', layout='centered' ) BASE_DIR = Path(__file__).resolve().parent MODEL_PATH = BASE_DIR / 'linear_regression_model.pkl' FEATURES_PATH = BASE_DIR / 'feature_names.pkl' st.title('🌊 Flood Probability Predictor') st.write('Predicts **FloodProbability** using a trained **Linear Regression** model.') @st.cache_resource def load_assets(): if not MODEL_PATH.exists(): raise FileNotFoundError( f'Model not found: {MODEL_PATH.name}. Put it in the repo root (same folder as app.py).' ) if not FEATURES_PATH.exists(): raise FileNotFoundError( f'Feature list not found: {FEATURES_PATH.name}. Put it in the repo root (same folder as app.py).' ) model = joblib.load(MODEL_PATH) feature_names = joblib.load(FEATURES_PATH) return model, feature_names model, feature_names = load_assets() st.subheader('Enter feature values') with st.form('input_form'): cols = st.columns(2) inputs = {} for i, feat in enumerate(feature_names): col = cols[i % 2] # Most features in your dataset are small integers (0-19) inputs[feat] = col.number_input( feat, min_value=0.0, max_value=30.0, value=5.0, step=1.0 ) submitted = st.form_submit_button('Predict') if submitted: x_row = pd.DataFrame([inputs], columns=feature_names) pred = float(model.predict(x_row)[0]) # Optional safety clip (not required by Kaggle, but safe for probabilities) pred_clipped = float(np.clip(pred, 0.0, 1.0)) st.subheader('Prediction') st.metric('FloodProbability', f'{pred_clipped:.4f}') with st.expander('Show raw prediction (before clipping)'): st.write(pred) with st.expander('Show input dataframe'): st.dataframe(x_row)