app.py

import streamlit as st
import requests
import json
import pandas as pd
import plotly.express as px
import plotly.graph_objects as go
from datetime import datetime
import time

# Page configuration
st.set_page_config(
    page_title="SuperKart Sales Forecasting",
    page_icon="🛒",
    layout="wide",
    initial_sidebar_state="expanded"
)

# Custom CSS for better styling
st.markdown("""
<style>
    .main-header {
        font-size: 3rem;
        color: #1f77b4;
        text-align: center;
        margin-bottom: 2rem;
    }
    .metric-card {
        background-color: #f0f2f6;
        padding: 1rem;
        border-radius: 0.5rem;
        border-left: 4px solid #1f77b4;
    }
    .prediction-result {
        font-size: 2rem;
        font-weight: bold;
        color: #2e8b57;
        text-align: center;
        padding: 1rem;
        background-color: #f0fff0;
        border-radius: 0.5rem;
        border: 2px solid #2e8b57;
    }
    .error-message {
        color: #dc3545;
        background-color: #f8d7da;
        padding: 1rem;
        border-radius: 0.5rem;
        border: 1px solid #f5c6cb;
    }
</style>
""", unsafe_allow_html=True)

# API Configuration
API_BASE_URL = "https://your-backend-api-url.hf.space"  # Replace with your actual API URL

def check_api_health():
    """Check if the API is healthy and accessible."""
    try:
        response = requests.get(f"{API_BASE_URL}/health", timeout=10)
        return response.status_code == 200
    except:
        return False

def make_prediction(data):
    """Make a single prediction using the API."""
    try:
        response = requests.post(
            f"{API_BASE_URL}/predict",
            json=data,
            headers={"Content-Type": "application/json"},
            timeout=30
        )
        
        if response.status_code == 200:
            return response.json(), None
        else:
            return None, f"API Error: {response.status_code} - {response.text}"
            
    except requests.exceptions.Timeout:
        return None, "Request timeout. Please try again."
    except requests.exceptions.ConnectionError:
        return None, "Cannot connect to API. Please check your internet connection."
    except Exception as e:
        return None, f"Unexpected error: {str(e)}"

def make_batch_prediction(data_list):
    """Make batch predictions using the API."""
    try:
        response = requests.post(
            f"{API_BASE_URL}/batch_predict",
            json=data_list,
            headers={"Content-Type": "application/json"},
            timeout=60
        )
        
        if response.status_code == 200:
            return response.json(), None
        else:
            return None, f"API Error: {response.status_code} - {response.text}"
            
    except requests.exceptions.Timeout:
        return None, "Request timeout. Please try again."
    except requests.exceptions.ConnectionError:
        return None, "Cannot connect to API. Please check your internet connection."
    except Exception as e:
        return None, f"Unexpected error: {str(e)}"

def main():
    """Main Streamlit application."""
    
    # Main header
    st.markdown('<h1 class="main-header">🛒 SuperKart Sales Forecasting</h1>', unsafe_allow_html=True)
    st.markdown("### AI-Powered Sales Prediction for Retail Excellence")
    
    # Sidebar for navigation
    st.sidebar.title("🎯 Navigation")
    app_mode = st.sidebar.selectbox("Choose the app mode",
                                   ["Single Prediction", "Batch Prediction", "Analytics Dashboard", "API Status"])
    
    # API Health Check
    with st.sidebar:
        st.markdown("---")
        st.subheader("🔧 API Status")
        
        if st.button("Check API Health"):
            with st.spinner("Checking API..."):
                if check_api_health():
                    st.success("✅ API is healthy")
                else:
                    st.error("❌ API is not accessible")
        
        st.markdown("---")
        st.markdown("""
        **Features:**
        - 🎯 Single Prediction
        - 📊 Batch Predictions
        - 📈 Analytics Dashboard
        - 🔍 Real-time Validation
        """)
    
    # Main content based on selected mode
    if app_mode == "Single Prediction":
        single_prediction_page()
    elif app_mode == "Batch Prediction":
        batch_prediction_page()
    elif app_mode == "Analytics Dashboard":
        analytics_dashboard_page()
    elif app_mode == "API Status":
        api_status_page()

def single_prediction_page():
    """Single prediction interface."""
    
    st.header("🎯 Single Sales Prediction")
    st.markdown("Enter product and store details to get an instant sales forecast.")
    
    # Create input form
    with st.form("prediction_form"):
        col1, col2, col3 = st.columns(3)
        
        with col1:
            st.subheader("📦 Product Details")
            product_weight = st.number_input(
                "Product Weight (kg)",
                min_value=0.1,
                max_value=50.0,
                value=12.5,
                step=0.1,
                help="Weight of the product in kilograms"
            )
            
            product_sugar_content = st.selectbox(
                "Sugar Content",
                ["Low Sugar", "Regular", "No Sugar"],
                help="Sugar content category of the product"
            )
            
            product_allocated_area = st.number_input(
                "Allocated Display Area",
                min_value=0.001,
                max_value=1.0,
                value=0.1,
                step=0.001,
                format="%.3f",
                help="Ratio of allocated display area (0-1)"
            )
            
            product_type = st.selectbox(
                "Product Type",
                [
                    "Fruits and Vegetables", "Snack Foods", "Household", "Frozen Foods",
                    "Dairy", "Canned", "Baking Goods", "Health and Hygiene", "Meat",
                    "Soft Drinks", "Hard Drinks", "Starchy Foods", "Breakfast",
                    "Seafood", "Bread", "Others"
                ],
                help="Category of the product"
            )
            
            product_mrp = st.number_input(
                "Maximum Retail Price (₹)",
                min_value=1.0,
                max_value=500.0,
                value=150.0,
                step=1.0,
                help="Maximum retail price in rupees"
            )
        
        with col2:
            st.subheader("🏪 Store Details")
            store_size = st.selectbox(
                "Store Size",
                ["Small", "Medium", "High"],
                index=1,
                help="Size category of the store"
            )
            
            store_location_city_type = st.selectbox(
                "City Type",
                ["Tier 1", "Tier 2", "Tier 3"],
                index=1,
                help="Tier classification of the city"
            )
            
            store_type = st.selectbox(
                "Store Type",
                ["Departmental Store", "Supermarket Type1", "Supermarket Type2", "Food Mart"],
                index=2,
                help="Type/format of the store"
            )
            
            store_age = st.number_input(
                "Store Age (years)",
                min_value=0,
                max_value=50,
                value=15,
                step=1,
                help="Age of the store in years"
            )
        
        with col3:
            st.subheader("📊 Prediction Summary")
            st.markdown("""
            **Input Validation:**
            - All fields are required
            - Weights: 0.1 - 50 kg
            - Display Area: 0.001 - 1.0
            - MRP: ₹1 - ₹500
            - Store Age: 0 - 50 years
            """)
            
            st.markdown("---")
            st.markdown("**Business Context:**")
            st.markdown("This prediction helps with:")
            st.markdown("- Inventory planning")
            st.markdown("- Revenue forecasting")
            st.markdown("- Store optimization")
            st.markdown("- Regional strategy")
        
        # Submit button
        submitted = st.form_submit_button("🚀 Predict Sales", use_container_width=True)
        
        if submitted:
            # Prepare data for API
            prediction_data = {
                "Product_Weight": product_weight,
                "Product_Sugar_Content": product_sugar_content,
                "Product_Allocated_Area": product_allocated_area,
                "Product_Type": product_type,
                "Product_MRP": product_mrp,
                "Store_Size": store_size,
                "Store_Location_City_Type": store_location_city_type,
                "Store_Type": store_type,
                "Store_Age": store_age
            }
            
            # Make prediction
            with st.spinner("🔮 Generating prediction..."):
                result, error = make_prediction(prediction_data)
                
                if result:
                    prediction = result["prediction"]
                    
                    # Display result
                    st.success("✅ Prediction Generated Successfully!")
                    
                    # Main prediction result
                    st.markdown(
                        f'<div class="prediction-result">💰 Predicted Sales: ₹{prediction:,.2f}</div>',
                        unsafe_allow_html=True
                    )
                    
                    # Additional insights
                    col1, col2, col3 = st.columns(3)
                    
                    with col1:
                        st.metric(
                            "Daily Revenue",
                            f"₹{prediction:,.2f}",
                            delta=f"{prediction*0.1:,.2f}",
                            delta_color="normal"
                        )
                    
                    with col2:
                        monthly_estimate = prediction * 30
                        st.metric(
                            "Monthly Estimate",
                            f"₹{monthly_estimate:,.2f}",
                            delta="Projected",
                            delta_color="off"
                        )
                    
                    with col3:
                        annual_estimate = prediction * 365
                        st.metric(
                            "Annual Potential",
                            f"₹{annual_estimate:,.2f}",
                            delta="Estimated",
                            delta_color="off"
                        )
                    
                    # Business recommendations
                    st.markdown("### 💡 Business Insights")
                    
                    if prediction > 4000:
                        st.success("🎯 **High Performance Expected** - This product-store combination shows excellent potential!")
                    elif prediction > 2500:
                        st.info("📈 **Good Performance Expected** - Solid sales potential with room for optimization.")
                    else:
                        st.warning("⚠️ **Moderate Performance Expected** - Consider promotional strategies or product mix optimization.")
                    
                    # Performance category analysis
                    if store_location_city_type == "Tier 1" and store_type == "Departmental Store":
                        st.info("🏆 **Premium Market Position** - Tier 1 Departmental Store typically shows highest performance.")
                    
                    if product_weight > 15:
                        st.info("📦 **Heavy Product Advantage** - Higher weight products tend to generate more sales.")
                    
                    if product_mrp > 200:
                        st.info("💎 **Premium Product** - High MRP products often indicate better margins.")
                
                else:
                    st.error(f"❌ Prediction Failed: {error}")
                    st.markdown("""
                    **Troubleshooting Steps:**
                    1. Check your internet connection
                    2. Verify API URL in the sidebar
                    3. Ensure all input values are within valid ranges
                    4. Try again in a few moments
                    """)

def batch_prediction_page():
    """Batch prediction interface."""
    
    st.header("📊 Batch Sales Prediction")
    st.markdown("Upload a CSV file or enter multiple records for bulk predictions.")
    
    # Option selection
    batch_option = st.radio(
        "Choose input method:",
        ["Upload CSV File", "Manual Entry"]
    )
    
    if batch_option == "Upload CSV File":
        st.subheader("📁 Upload CSV File")
        
        # File upload
        uploaded_file = st.file_uploader(
            "Choose a CSV file",
            type="csv",
            help="Upload a CSV file with the required columns"
        )
        
        # Show required format
        with st.expander("📋 Required CSV Format"):
            sample_df = pd.DataFrame({
                "Product_Weight": [12.5, 16.2, 8.9],
                "Product_Sugar_Content": ["Low Sugar", "Regular", "No Sugar"],
                "Product_Allocated_Area": [0.1, 0.15, 0.05],
                "Product_Type": ["Fruits and Vegetables", "Dairy", "Snack Foods"],
                "Product_MRP": [150.0, 180.0, 95.0],
                "Store_Size": ["Medium", "High", "Small"],
                "Store_Location_City_Type": ["Tier 2", "Tier 1", "Tier 3"],
                "Store_Type": ["Supermarket Type2", "Departmental Store", "Food Mart"],
                "Store_Age": [15, 20, 8]
            })
            st.dataframe(sample_df)
        
        if uploaded_file is not None:
            try:
                # Read CSV
                df = pd.read_csv(uploaded_file)
                st.success(f"✅ File uploaded successfully! {len(df)} records found.")
                
                # Show preview
                st.subheader("📋 Data Preview")
                st.dataframe(df.head())
                
                # Validate columns
                required_columns = [
                    "Product_Weight", "Product_Sugar_Content", "Product_Allocated_Area",
                    "Product_Type", "Product_MRP", "Store_Size", 
                    "Store_Location_City_Type", "Store_Type", "Store_Age"
                ]
                
                missing_columns = [col for col in required_columns if col not in df.columns]
                
                if missing_columns:
                    st.error(f"❌ Missing required columns: {missing_columns}")
                else:
                    st.success("✅ All required columns found!")
                    
                    if st.button("🚀 Generate Batch Predictions"):
                        # Convert DataFrame to list of dictionaries
                        data_list = df.to_dict('records')
                        
                        with st.spinner(f"🔮 Generating predictions for {len(data_list)} records..."):
                            result, error = make_batch_prediction(data_list)
                            
                            if result:
                                predictions = result["predictions"]
                                
                                # Add predictions to DataFrame
                                df_results = df.copy()
                                df_results["Predicted_Sales"] = predictions
                                
                                # Display results
                                st.success("✅ Batch Predictions Generated Successfully!")
                                
                                # Summary metrics
                                col1, col2, col3, col4 = st.columns(4)
                                
                                with col1:
                                    st.metric("Total Records", len(predictions))
                                
                                with col2:
                                    successful = len([p for p in predictions if p is not None])
                                    st.metric("Successful", successful)
                                
                                with col3:
                                    avg_prediction = sum([p for p in predictions if p is not None]) / successful
                                    st.metric("Average Sales", f"₹{avg_prediction:,.2f}")
                                
                                with col4:
                                    total_predicted = sum([p for p in predictions if p is not None])
                                    st.metric("Total Predicted", f"₹{total_predicted:,.2f}")
                                
                                # Results table
                                st.subheader("📊 Prediction Results")
                                st.dataframe(df_results)
                                
                                # Download results
                                csv_results = df_results.to_csv(index=False)
                                st.download_button(
                                    "📥 Download Results",
                                    csv_results,
                                    "superkart_predictions.csv",
                                    "text/csv"
                                )
                                
                                # Visualization
                                if successful > 0:
                                    st.subheader("📈 Prediction Analysis")
                                    
                                    # Distribution plot
                                    valid_predictions = [p for p in predictions if p is not None]
                                    fig = px.histogram(
                                        x=valid_predictions,
                                        title="Distribution of Predicted Sales",
                                        labels={"x": "Predicted Sales (₹)", "y": "Frequency"}
                                    )
                                    st.plotly_chart(fig, use_container_width=True)
                            
                            else:
                                st.error(f"❌ Batch Prediction Failed: {error}")
                
            except Exception as e:
                st.error(f"❌ Error reading file: {str(e)}")
    
    else:  # Manual Entry
        st.subheader("✏️ Manual Entry")
        st.markdown("Add multiple records manually for batch prediction.")
        
        # Initialize session state for manual entries
        if "manual_entries" not in st.session_state:
            st.session_state.manual_entries = []
        
        # Add new entry form
        with st.expander("➕ Add New Entry"):
            with st.form("manual_entry_form"):
                col1, col2 = st.columns(2)
                
                with col1:
                    weight = st.number_input("Weight (kg)", 0.1, 50.0, 12.5, key="manual_weight")
                    sugar = st.selectbox("Sugar Content", ["Low Sugar", "Regular", "No Sugar"], key="manual_sugar")
                    area = st.number_input("Display Area", 0.001, 1.0, 0.1, key="manual_area")
                    product_type = st.selectbox("Product Type", [
                        "Fruits and Vegetables", "Snack Foods", "Household", "Frozen Foods",
                        "Dairy", "Canned", "Baking Goods", "Health and Hygiene"
                    ], key="manual_type")
                    mrp = st.number_input("MRP (₹)", 1.0, 500.0, 150.0, key="manual_mrp")
                
                with col2:
                    size = st.selectbox("Store Size", ["Small", "Medium", "High"], key="manual_size")
                    city = st.selectbox("City Type", ["Tier 1", "Tier 2", "Tier 3"], key="manual_city")
                    store_type = st.selectbox("Store Type", [
                        "Departmental Store", "Supermarket Type1", "Supermarket Type2", "Food Mart"
                    ], key="manual_store_type")
                    age = st.number_input("Store Age", 0, 50, 15, key="manual_age")
                
                if st.form_submit_button("➕ Add Entry"):
                    entry = {
                        "Product_Weight": weight,
                        "Product_Sugar_Content": sugar,
                        "Product_Allocated_Area": area,
                        "Product_Type": product_type,
                        "Product_MRP": mrp,
                        "Store_Size": size,
                        "Store_Location_City_Type": city,
                        "Store_Type": store_type,
                        "Store_Age": age
                    }
                    st.session_state.manual_entries.append(entry)
                    st.success("✅ Entry added!")
        
        # Display current entries
        if st.session_state.manual_entries:
            st.subheader(f"📝 Current Entries ({len(st.session_state.manual_entries)})")
            
            # Convert to DataFrame for display
            entries_df = pd.DataFrame(st.session_state.manual_entries)
            st.dataframe(entries_df)
            
            col1, col2 = st.columns(2)
            
            with col1:
                if st.button("🚀 Generate Predictions"):
                    with st.spinner("🔮 Generating predictions..."):
                        result, error = make_batch_prediction(st.session_state.manual_entries)
                        
                        if result:
                            predictions = result["predictions"]
                            entries_df["Predicted_Sales"] = predictions
                            
                            st.success("✅ Predictions Generated!")
                            st.dataframe(entries_df)
                            
                            # Download option
                            csv_data = entries_df.to_csv(index=False)
                            st.download_button(
                                "📥 Download Results",
                                csv_data,
                                "manual_predictions.csv",
                                "text/csv"
                            )
                        else:
                            st.error(f"❌ Prediction Failed: {error}")
            
            with col2:
                if st.button("🗑️ Clear All Entries"):
                    st.session_state.manual_entries = []
                    st.experimental_rerun()

def analytics_dashboard_page():
    """Analytics dashboard interface."""
    
    st.header("📈 Analytics Dashboard")
    st.markdown("Explore sales patterns and model insights.")
    
    # Mock analytics data for demonstration
    st.subheader("🎯 Model Performance Metrics")
    
    col1, col2, col3, col4 = st.columns(4)
    
    with col1:
        st.metric("Model Accuracy", "92.8%", "2.1%")
    
    with col2:
        st.metric("Avg Prediction Error", "₹249", "-₹15")
    
    with col3:
        st.metric("Total Predictions", "1,247", "156")
    
    with col4:
        st.metric("API Uptime", "99.2%", "0.3%")
    
    # Feature importance chart
    st.subheader("🎯 Feature Importance")
    
    feature_data = {
        "Feature": ["Product_Weight", "Product_MRP", "Store_Type", "City_Type", "Store_Size"],
        "Importance": [0.35, 0.28, 0.18, 0.12, 0.07]
    }
    
    fig = px.bar(
        x=feature_data["Importance"],
        y=feature_data["Feature"],
        orientation='h',
        title="Top 5 Most Important Features",
        labels={"x": "Importance Score", "y": "Features"}
    )
    fig.update_layout(yaxis={'categoryorder':'total ascending'})
    st.plotly_chart(fig, use_container_width=True)
    
    # Sample insights
    st.subheader("💡 Business Insights")
    
    insight_tabs = st.tabs(["Store Performance", "Product Analysis", "Regional Trends"])
    
    with insight_tabs[0]:
        st.markdown("""
        **Store Performance Insights:**
        - Departmental Stores show 40% higher sales on average
        - Medium-sized stores have the best cost-to-performance ratio
        - Tier 1 cities generate 2.8x more revenue than Tier 3
        """)
    
    with insight_tabs[1]:
        st.markdown("""
        **Product Analysis:**
        - Heavy products (>15kg) correlate with higher sales
        - Premium MRP products (>₹200) show better margins
        - Dairy and Frozen Foods are top performing categories
        """)
    
    with insight_tabs[2]:
        st.markdown("""
        **Regional Trends:**
        - Tier 1 cities: Focus on premium product mix
        - Tier 2 cities: Balanced approach with growth potential
        - Tier 3 cities: Price-sensitive, high-volume strategy
        """)

def api_status_page():
    """API status and configuration page."""
    
    global API_BASE_URL
    
    st.header("🔧 API Status & Configuration")
    
    # API URL configuration
    st.subheader("⚙️ API Configuration")
    
    current_url = st.text_input(
        "Backend API URL",
        value=API_BASE_URL,
        help="Enter your backend API URL"
    )
    
    if st.button("💾 Update API URL"):
        API_BASE_URL = current_url
        st.success("✅ API URL updated!")
    
    # Health check
    st.subheader("🏥 Health Check")
    
    if st.button("🔍 Check API Health"):
        with st.spinner("Checking API health..."):
            health_status = check_api_health()
            
            if health_status:
                st.success("✅ API is healthy and responsive!")
                
                # Try to get API info
                try:
                    response = requests.get(f"{API_BASE_URL}/", timeout=10)
                    if response.status_code == 200:
                        api_info = response.json()
                        st.json(api_info)
                except:
                    pass
            else:
                st.error("❌ API is not accessible")
                st.markdown("""
                **Troubleshooting:**
                1. Check if the API URL is correct
                2. Ensure the backend service is running
                3. Verify your internet connection
                4. Check if the API allows CORS requests
                """)
    
    # API documentation
    st.subheader("📚 API Documentation")
    
    st.markdown("""
    **Available Endpoints:**
    
    - `GET /` - API information and sample input
    - `GET /health` - Health check endpoint
    - `GET /model_info` - Model details and performance metrics
    - `POST /predict` - Single prediction endpoint
    - `POST /batch_predict` - Batch prediction endpoint
    
    **Sample Request Format:**
    ```json
    {
        "Product_Weight": 12.5,
        "Product_Sugar_Content": "Low Sugar",
        "Product_Allocated_Area": 0.15,
        "Product_Type": "Fruits and Vegetables",
        "Product_MRP": 150.0,
        "Store_Size": "Medium",
        "Store_Location_City_Type": "Tier 2",
        "Store_Type": "Supermarket Type2",
        "Store_Age": 15
    }
    ```
    """)

# Footer
def show_footer():
    """Show application footer."""
    st.markdown("---")
    st.markdown("""
    <div style='text-align: center; color: #666;'>
        <p>🛒 SuperKart Sales Forecasting System | Powered by AI & Machine Learning</p>
        <p>Built with Streamlit & Flask | © 2025 SuperKart Analytics Team</p>
    </div>
    """, unsafe_allow_html=True)

# Run the application
if __name__ == "__main__":
    main()
    show_footer()