import streamlit as st 
import pandas as pd 
import numpy as np 
from sklearn.model_selection import train_test_split
from sklearn.svm import SVR
from sklearn.metrics import mean_absolute_error

st.set_page_config(page_title="ETA Predictor", layout="wide")

st.markdown("""
    <style>
    .css-1cypcdb, .css-1p05t8e {
        font-size: 11px !important;
        color: #2f4f4f !important;
    }
    section[data-testid="stSidebar"] {
        background-color: #f0f2f6;
    }
    h1 {
        font-size: 20px;
        color: #003366;
    }
    h3 {
        font-size: 14px;
        color: #cc6600;
    }
    </style>
""", unsafe_allow_html=True)

st.title("🚗 ETA Predictor")
st.markdown("""
    <h5>This tool estimates travel time between two points in Hyderabad, taking into account variables like weather, traffic, weekday, and time of day. Adjust the values in the sidebar and click the button to get a real-time prediction of your Estimated Time of Arrival (ETA).</h5>
""", unsafe_allow_html=True)

df = pd.read_csv("hyderabad_eta_data (1).csv")

st.sidebar.header("Input Parameters")

start_lat = st.sidebar.slider("Start Latitude", -90.0, 90.0, 0.0, 0.1)
start_lon = st.sidebar.slider("Start Longitude", -180.0, 180.0, 0.0, 0.1)
end_lat = st.sidebar.slider("End Latitude", -90.0, 90.0, 0.0, 0.1)
end_lon = st.sidebar.slider("End Longitude", -180.0, 180.0, 0.0, 0.1)
distance = st.sidebar.slider("Distance (km)", 0.0, 700.0, 100.0, 1.0)

traffic = st.sidebar.selectbox("Traffic Density (1-10)", list(range(1, 11)))
weather = st.sidebar.selectbox("Weather", ["rainy", "clear", "foggy"])
week_day = st.sidebar.selectbox("Weekday", ["Sunday", "Monday", "Tuesday", "Wednesday", "Thursday", "Friday", "Saturday"])
hour = st.sidebar.slider("Hour (0-23)", 0, 23, 0, 1)

week_val = {"Sunday": 0, "Monday": 1, "Tuesday": 2, "Wednesday": 3, "Thursday": 4, "Friday": 5, "Saturday": 6}
weather_map = {"rainy": 0, "clear": 1, "foggy": 2}

week = week_val[week_day]
weather_val = weather_map[weather]

df["weather_condition"] = df["weather_condition"].map(weather_map)
X = df.drop("ETA", axis=1)
y = df["ETA"]
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=23)

sv_r = SVR(kernel='rbf', C=100, epsilon=0.1)
sv_r.fit(X_train, y_train)
y_pred = sv_r.predict(X_test)
mae = mean_absolute_error(y_pred, y_test)

features = np.array([[start_lat, start_lon, end_lat, end_lon, distance, weather_val, traffic, week, hour]])
estimated_arrival = sv_r.predict(features)[0]

st.subheader("Model Performance")
st.markdown(f"<p style='font-size:12px;'>Mean Absolute Error: <b>{mae:.2f} hrs</b></p>", unsafe_allow_html=True)

if st.button("Predict ETA"):
    st.subheader("Predicted ETA")
    st.markdown(f"<h2 style='color:brown; font-size:20px;'><strong>{estimated_arrival:.2f} hours</strong></h2>", unsafe_allow_html=True)