Spaces:
Sleeping
Sleeping
| # Enhanced Streamlit Web App for SuperKart Sales Forecasting | |
| import streamlit as st | |
| import requests | |
| import numpy as np | |
| import pandas as pd | |
| import plotly.express as px | |
| import plotly.graph_objects as go | |
| from datetime import datetime | |
| # 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 { | |
| background: linear-gradient(90deg, #667eea 0%, #764ba2 100%); | |
| padding: 2rem; | |
| border-radius: 10px; | |
| margin-bottom: 2rem; | |
| text-align: center; | |
| color: white; | |
| } | |
| .metric-card { | |
| background: #f8f9fa; | |
| padding: 1rem; | |
| border-radius: 8px; | |
| border: 1px solid #e9ecef; | |
| margin: 0.5rem 0; | |
| } | |
| .prediction-result { | |
| background: #d4edda; | |
| border: 1px solid #c3e6cb; | |
| padding: 1.5rem; | |
| border-radius: 10px; | |
| margin: 1rem 0; | |
| text-align: center; | |
| } | |
| .sidebar-content { | |
| background: #f8f9fa; | |
| padding: 1rem; | |
| border-radius: 8px; | |
| margin-bottom: 1rem; | |
| } | |
| </style> | |
| """, unsafe_allow_html=True) | |
| # Header | |
| st.markdown(""" | |
| <div class="main-header"> | |
| <h1>🛒 SuperKart Sales Forecasting System</h1> | |
| <p>AI-Powered Revenue Prediction for Retail Excellence</p> | |
| </div> | |
| """, unsafe_allow_html=True) | |
| # Sidebar for inputs | |
| st.sidebar.header("📊 Product & Store Configuration") | |
| # Add logo to sidebar | |
| st.sidebar.image("https://cdn-icons-png.flaticon.com/512/2331/2331970.png", width=250, caption="SuperKart Analytics") | |
| # Instructions in sidebar | |
| st.sidebar.markdown(""" | |
| ### 🔍 How to Use: | |
| 1. **Configure** product attributes | |
| 2. **Select** store characteristics | |
| 3. **Click** Predict to get forecast | |
| 4. **Analyze** results and insights | |
| *All predictions are in USD ($)* | |
| """) | |
| # Create tabs for better organization | |
| tab1, tab2, tab3 = st.tabs(["📈 Prediction", "📊 Analytics", "ℹ️ About"]) | |
| with tab1: | |
| # Create two columns for inputs | |
| col1, col2 = st.columns(2) | |
| with col1: | |
| st.subheader("🏷️ Product Attributes") | |
| Product_Weight = st.number_input( | |
| "Product Weight (oz)", | |
| min_value=0.0, | |
| max_value=50.0, | |
| value=12.66, | |
| step=0.1, | |
| help="Weight of the product in ounces" | |
| ) | |
| Product_Sugar_Content = st.selectbox( | |
| "Sugar Content Level", | |
| ["Low Sugar", "Regular", "No Sugar"], | |
| help="Sugar content classification of the product" | |
| ) | |
| Product_Allocated_Area = st.number_input( | |
| "Allocated Shelf Area (sq. in.)", | |
| min_value=0.0, | |
| max_value=500.0, | |
| value=100.0, | |
| step=1.0, | |
| help="Shelf space allocated to the product" | |
| ) | |
| Product_MRP = st.number_input( | |
| "Maximum Retail Price (USD)", | |
| min_value=0.0, | |
| max_value=1000.0, | |
| value=150.0, | |
| step=0.5, | |
| help="Maximum retail price of the product" | |
| ) | |
| Product_Type_Category = st.selectbox( | |
| "Product Category", | |
| ["Perishables", "Non Perishables"], | |
| help="Whether the product is perishable or not" | |
| ) | |
| with col2: | |
| st.subheader("🏪 Store Characteristics") | |
| Store_Size = st.selectbox( | |
| "Store Size Category", | |
| ["Small", "Medium", "High"], | |
| help="Size classification of the store" | |
| ) | |
| Store_Location_City_Type = st.selectbox( | |
| "City Classification", | |
| ["Tier 1", "Tier 2", "Tier 3"], | |
| help="Urban development tier of store location" | |
| ) | |
| Store_Type = st.selectbox( | |
| "Store Format", | |
| ["Supermarket Type1", "Supermarket Type2", "Departmental Store", "Food Mart"], | |
| help="Type and format of the retail store" | |
| ) | |
| Store_Age_Years = st.slider( | |
| "Store Age (years)", | |
| min_value=0, | |
| max_value=30, | |
| value=10, | |
| help="Number of years the store has been operational" | |
| ) | |
| # Display current configuration | |
| st.subheader("🔧 Current Configuration") | |
| config_col1, config_col2, config_col3 = st.columns(3) | |
| with config_col1: | |
| st.metric("Product Weight", f"{Product_Weight} oz") | |
| st.metric("Allocated Area", f"{Product_Allocated_Area} sq.in") | |
| with config_col2: | |
| st.metric("MRP", f"${Product_MRP}") | |
| st.metric("Store Age", f"{Store_Age_Years} years") | |
| with config_col3: | |
| st.metric("Sugar Content", Product_Sugar_Content) | |
| st.metric("Store Type", Store_Type) | |
| # Apply log1p transform (must match backend model training) | |
| Product_Allocated_Area_Log = np.log1p(Product_Allocated_Area) | |
| # Prepare JSON payload for the backend | |
| product_data = { | |
| "Product_Weight": str(Product_Weight), | |
| "Product_Sugar_Content": Product_Sugar_Content, | |
| "Product_Allocated_Area": str(Product_Allocated_Area), | |
| "Product_MRP": str(Product_MRP), | |
| "Store_Size": Store_Size, | |
| "Store_Location_City_Type": Store_Location_City_Type, | |
| "Store_Type": Store_Type, | |
| "Store_Age_Years": str(Store_Age_Years), | |
| "Product_Type_Category": Product_Type_Category | |
| } | |
| # Prediction section | |
| st.subheader("🎯 Sales Prediction") | |
| predict_col1, predict_col2, predict_col3 = st.columns([1, 2, 1]) | |
| with predict_col2: | |
| if st.button("🚀 Generate Prediction", type='primary', use_container_width=True): | |
| with st.spinner('🔄 Analyzing data and generating prediction...'): | |
| try: | |
| response = requests.post( | |
| "https://arnavarpit-super-kart-backend.hf.space/v1/predict", | |
| json=product_data, | |
| timeout=30 | |
| ) | |
| if response.status_code == 200: | |
| result = response.json() | |
| predicted_sales = float(result["Predicted_Sales"]) | |
| # Display result with enhanced styling | |
| st.markdown(f""" | |
| <div class="prediction-result"> | |
| <h2>📈 Prediction Results</h2> | |
| <h1 style="color: #28a745; font-size: 3rem;">${predicted_sales:,.2f}</h1> | |
| <p style="font-size: 1.2rem;">Estimated Monthly Sales Revenue</p> | |
| <p><em>Generated on {datetime.now().strftime('%B %d, %Y at %I:%M %p')}</em></p> | |
| </div> | |
| """, unsafe_allow_html=True) | |
| # Additional insights | |
| st.subheader("💡 Insights & Recommendations") | |
| insights_col1, insights_col2 = st.columns(2) | |
| with insights_col1: | |
| if predicted_sales > 2000: | |
| st.success("🎉 **High Revenue Potential!** This configuration shows excellent sales prospects.") | |
| elif predicted_sales > 1000: | |
| st.info("📊 **Moderate Revenue** Expected performance within normal range.") | |
| else: | |
| st.warning("⚠️ **Lower Revenue Forecast** Consider optimizing product or store factors.") | |
| with insights_col2: | |
| # Performance category | |
| if predicted_sales > 2500: | |
| category = "⭐ Excellent" | |
| elif predicted_sales > 1500: | |
| category = "👍 Good" | |
| elif predicted_sales > 800: | |
| category = "📈 Average" | |
| else: | |
| category = "📉 Below Average" | |
| st.metric("Performance Category", category) | |
| # Store results in session state for analytics | |
| if 'predictions' not in st.session_state: | |
| st.session_state.predictions = [] | |
| st.session_state.predictions.append({ | |
| 'timestamp': datetime.now(), | |
| 'predicted_sales': predicted_sales, | |
| 'configuration': product_data.copy() | |
| }) | |
| else: | |
| st.error(f"❌ **API Error** (Status: {response.status_code})") | |
| st.error("Please verify input values or try again later.") | |
| except requests.exceptions.Timeout: | |
| st.error("⏰ **Request Timeout** - The prediction service is taking too long to respond.") | |
| except requests.exceptions.ConnectionError: | |
| st.error("🌐 **Connection Error** - Unable to reach the prediction service.") | |
| except Exception as e: | |
| st.error(f"⚠️ **Unexpected Error**: {str(e)}") | |
| with tab2: | |
| st.subheader("📊 Analytics Dashboard") | |
| if 'predictions' in st.session_state and st.session_state.predictions: | |
| df = pd.DataFrame([ | |
| { | |
| 'Timestamp': pred['timestamp'], | |
| 'Predicted Sales': pred['predicted_sales'], | |
| 'Product Weight': float(pred['configuration']['Product_Weight']), | |
| 'MRP': float(pred['configuration']['Product_MRP']), | |
| 'Store Age': int(pred['configuration']['Store_Age_Years']) | |
| } | |
| for pred in st.session_state.predictions | |
| ]) | |
| # Display recent predictions | |
| st.write("### Recent Predictions") | |
| st.dataframe(df.sort_values('Timestamp', ascending=False), use_container_width=True) | |
| # Create visualizations | |
| if len(df) > 1: | |
| col1, col2 = st.columns(2) | |
| with col1: | |
| fig1 = px.line(df, x='Timestamp', y='Predicted Sales', | |
| title='Sales Predictions Over Time') | |
| st.plotly_chart(fig1, use_container_width=True) | |
| with col2: | |
| fig2 = px.scatter(df, x='MRP', y='Predicted Sales', | |
| size='Product Weight', title='Sales vs MRP') | |
| st.plotly_chart(fig2, use_container_width=True) | |
| # Clear history button | |
| if st.button("🗑️ Clear Prediction History"): | |
| st.session_state.predictions = [] | |
| st.rerun() | |
| else: | |
| st.info("📝 Generate some predictions first to see analytics here!") | |
| with tab3: | |
| st.subheader("ℹ️ About This Application") | |
| st.markdown(""" | |
| ### 🎯 Purpose | |
| This application uses machine learning to predict monthly sales revenue for SuperKart retail stores based on product and store characteristics. | |
| ### 🔧 How It Works | |
| 1. **Data Input**: Enter product attributes and store characteristics | |
| 2. **ML Processing**: Advanced algorithms analyze the input features | |
| 3. **Prediction**: Generate accurate sales revenue forecasts | |
| 4. **Insights**: Receive actionable recommendations | |
| ### 📊 Key Features | |
| - **Real-time Predictions**: Instant ML-powered forecasting | |
| - **Interactive Interface**: User-friendly input controls | |
| - **Visual Analytics**: Charts and trend analysis | |
| - **Performance Insights**: Automated recommendations | |
| ### 🛠️ Technical Details | |
| - **Backend**: HuggingFace Spaces API | |
| - **Model**: Trained on historical SuperKart sales data | |
| - **Accuracy**: Optimized for retail revenue prediction | |
| - **Response Time**: < 3 seconds average | |
| ### 📝 Usage Notes | |
| - All monetary values are in USD ($) | |
| - Predictions are for monthly revenue estimates | |
| - Results may vary based on market conditions | |
| - For best accuracy, use realistic input values | |
| """) | |
| # Add download button for prediction history | |
| if 'predictions' in st.session_state and st.session_state.predictions: | |
| df_download = pd.DataFrame(st.session_state.predictions) | |
| csv = df_download.to_csv(index=False) | |
| st.download_button( | |
| label="📥 Download Prediction History", | |
| data=csv, | |
| file_name=f"superkart_predictions_{datetime.now().strftime('%Y%m%d')}.csv", | |
| mime="text/csv" | |
| ) | |