ui.py

import streamlit as st
import pandas as pd
import joblib

# Load trained model
model = joblib.load("travel_time_model.pkl")

# Page title
st.title("Travel Time Prediction System")
st.write("Predict travel time using traffic conditions")

st.divider()

# User Inputs
distance = st.number_input("Distance (in km)", min_value=0.1, value=5.0)
speed = st.number_input("Average Speed (km/h)", min_value=1.0, value=25.0)

time_of_day = st.selectbox("Time of Day",["Morning", "Afternoon", "Evening", "Night"])
traffic_density = st.selectbox("Traffic Density Level", ["Low", "Medium", "High"])

# Encoding
time_of_day_map = {"Morning": 0,"Afternoon": 1,"Evening": 2,"Night": 3}
traffic_density_map = {"Low": 0,"Medium": 1,"High": 2}

# Prediction Button
if st.button("Predict Travel Time"):
    input_df = pd.DataFrame([{
        "distance_km": distance,
        "average_speed_kmph": speed,
        "time_of_day": time_of_day_map[time_of_day],
        "traffic_density_level": traffic_density_map[traffic_density]
    }])
    prediction = model.predict(input_df)
    st.success(f"Estimated Travel Time: {prediction[0]:.2f} minutes")

st.divider()

st.caption( "Machine Learning Model: Random Forest Regressor | Dataset: Delhi Traffic Data")