import streamlit as st
import pandas as pd
import numpy as np
import pickle



import base64

def get_base64_of_image(image_path):
    with open(image_path, "rb") as img_file:
        data = img_file.read()
    return base64.b64encode(data).decode()


def set_background_local(image_path):
    base64_img = get_base64_of_image(image_path)
    st.markdown(f"""
        <style>
        .stApp {{
            background-image: url("data:image/jpg;base64,{base64_img}");
            background-size: cover;
            background-repeat: no-repeat;
            background-attachment: fixed;
            background-position: center;
        }}
        .overlay {{
            position: fixed;
            top: 0;
            left: 0;
            width: 100vw;
            height: 100vh;
            background-color: rgba(0, 0, 0, 0.2);
            z-index: 0;
        }}
        </style>
        <div class="overlay"></div>
    """, unsafe_allow_html=True)





def enable_glassmorphism():
    st.markdown("""
    <style>
    section.main > div {{
        backdrop-filter: blur(100px);
        background-color: rgba(255, 0, 0, 0.08);
        border-radius: 16px;
        padding: 1.5rem;
    }}
    </style>
    """, unsafe_allow_html=True)
















st.set_page_config(page_title="Rain Prediction App", page_icon="🌧️", layout="centered")
# Inject CSS for raining effect



set_background_local("filip-zrnzevic-_EMkxLdko9k-unsplash.jpg")
enable_glassmorphism()
# Load model
with open("rain.pkl", "rb") as file:
    model = pickle.load(file)

# Title
st.markdown("<h1 style='text-align: center; color: #ffffff;'>🌧️ Rain Prediction App</h1>", unsafe_allow_html=True)
st.markdown("<p style='text-align: center;'>Enter today’s weather details below to predict if it will rain tomorrow.</p>", unsafe_allow_html=True)

# Sidebar icon
st.sidebar.image("https://cdn-icons-png.flaticon.com/512/1163/1163624.png", width=100)
st.sidebar.markdown("### ⛅ About")
st.sidebar.info("This app uses a trained machine learning model to predict the likelihood of rainfall tomorrow based on various weather parameters.")

# Input Form
with st.form("rain_form"):
    st.markdown("### 🌡️ Weather Inputs")

    col1, col2 = st.columns(2)

    with col1:
        min_temp = st.number_input("Min Temperature (°C)", value=0.0)
        rainfall = st.number_input("Rainfall (mm)", value=0.0)
        sunshine = st.number_input("Sunshine (hrs)", value=0.0)
        wind_speed_9am = st.number_input("Wind Speed at 9 AM (km/h)", value=0.0)
        humidity_9am = st.slider("Humidity at 9 AM (%)", 0, 100, 50)
        pressure_9am = st.number_input("Pressure at 9 AM (hPa)", value=1010.0)
        cloud_9am = st.slider("Cloud at 9 AM (0-9)", 0, 9, 4)
        temp_9am = st.number_input("Temperature at 9 AM (°C)", value=15.0)

    with col2:
        max_temp = st.number_input("Max Temperature (°C)", value=0.0)
        evaporation = st.number_input("Evaporation (mm)", value=0.0)
        wind_gust_speed = st.number_input("Wind Gust Speed (km/h)", value=0.0)
        wind_speed_3pm = st.number_input("Wind Speed at 3 PM (km/h)", value=0.0)
        humidity_3pm = st.slider("Humidity at 3 PM (%)", 0, 100, 50)
        pressure_3pm = st.number_input("Pressure at 3 PM (hPa)", value=1010.0)
        cloud_3pm = st.slider("Cloud at 3 PM (0-9)", 0, 9, 4)
        temp_3pm = st.number_input("Temperature at 3 PM (°C)", value=15.0)

    rain_today = st.selectbox("🌧️ Did it rain today?", ["No", "Yes"])

    submitted = st.form_submit_button("🔍 Predict")

# Prediction
if submitted:
    input_data = pd.DataFrame([[
        min_temp, max_temp, rainfall, evaporation, sunshine,
        wind_gust_speed, wind_speed_9am, wind_speed_3pm,
        humidity_9am, humidity_3pm, pressure_9am, pressure_3pm,
        cloud_9am, cloud_3pm, temp_9am, temp_3pm, rain_today
    ]], columns=[
        'MinTemp', 'MaxTemp', 'Rainfall', 'Evaporation', 'Sunshine',
        'WindGustSpeed', 'WindSpeed9am', 'WindSpeed3pm',
        'Humidity9am', 'Humidity3pm', 'Pressure9am', 'Pressure3pm',
        'Cloud9am', 'Cloud3pm', 'Temp9am', 'Temp3pm', 'RainToday'
    ])

    prediction = model.predict(input_data)[0]
    proba = model.predict_proba(input_data)[0][1]
    if prediction == 'Yes':
        if proba < 0.5:
            
            st.success("☔ It **will rain tomorrow.**")

        else:
            
            st.success("☔ It **will rain tomorrow.**")

    else:
        st.info("☀️ It **won't rain tomorrow.**")
    st.markdown("### 🌦️ Prediction Result")


    st.markdown(f"### 🔎 Confidence: `{proba:.2%}`")

    st.progress(proba)