Code Editor for app.py

import streamlit as st
import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression
from datetime import datetime, timedelta

st.title("Power Consumption Prediction")

# Input URL
url = st.text_input(
    "Enter CSV URL",
    value="/kaggle/input/electric-power-consumption-data-set",
    key="url_input"
)

@st.cache_data
def load_data(url):
    try:
        # Add error handling for the URL
        if url.startswith("/kaggle"):
            st.error("Please provide a valid URL. The Kaggle path is not accessible directly.")
            return None
        
        df = pd.read_csv(url, sep=';', parse_dates=['Date Time'])
        return df
    except Exception as e:
        st.error(f"Error loading data: {str(e)}")
        return None

# Load and prepare data
if url:
    df = load_data(url)
    
    if df is not None:
        st.subheader("Data Preview")
        st.dataframe(df.head())
        
        # Prepare data for prediction
        df['Hour'] = df['Date Time'].dt.hour
        df['DayOfWeek'] = df['Date Time'].dt.dayofweek
        df['Month'] = df['Date Time'].dt.month
        
        # Features for prediction
        features = ['Hour', 'DayOfWeek', 'Month']
        X = df[features]
        y = df['Global_active_power']
        
        # Train model
        model = LinearRegression()
        model.fit(X, y)
        
        # Input for prediction
        st.subheader("Enter Current Power Consumption Details")
        
        current_hour = st.number_input("Hour (0-23)", min_value=0, max_value=23, value=datetime.now().hour)
        current_day = st.number_input("Day of Week (0=Monday, 6=Sunday)", min_value=0, max_value=6, value=datetime.now().weekday())
        current_month = st.number_input("Month (1-12)", min_value=1, max_value=12, value=datetime.now().month)
        
        if st.button("Predict Next Year's Consumption"):
            # Create input data for prediction
            input_data = pd.DataFrame({
                'Hour': [current_hour],
                'DayOfWeek': [current_day],
                'Month': [current_month]
            })
            
            # Make prediction
            prediction = model.predict(input_data)[0]
            
            st.success(f"Predicted power consumption for next year at this time: {prediction:.2f} kilowatts")
            
            # Show additional insights
            st.subheader("Historical Statistics")
            stats_df = df.groupby('Month')['Global_active_power'].agg(['mean', 'min', 'max']).round(2)
            stats_df.columns = ['Average kW', 'Minimum kW', 'Maximum kW']
            st.write("Monthly Power Consumption Statistics:")
            st.table(stats_df)
            
            # Plot monthly trends
            st.subheader("Monthly Trends")
            monthly_avg = df.groupby('Month')['Global_active_power'].mean()
            st.line_chart(monthly_avg)