app.py

import streamlit as st
import pandas as pd
import requests
from datetime import datetime

# Set the title of the Streamlit app
st.title("Retail Product Sales Prediction")

# Section for single prediction
st.subheader("Single Product-Store Prediction")

# Create two columns for better layout
col1, col2 = st.columns(2)

with col1:
    # Product Features
    st.markdown("**Product Details**")
    product_weight = st.slider("Product Weight (kg)", 4.0, 22.0, 12.66, 0.01)
    product_sugar = st.selectbox("Sugar Content", ["Low", "Medium", "High"])
    product_area = st.slider("Allocated Area (sqm)", 0.004, 0.3, 0.05, 0.001)
    product_type = st.selectbox("Product Type", [
        "Frozen Foods", "Dairy", "Canned", "Baking Goods",
        "Health and Hygiene", "Snack Foods", "Meat", "Household"
    ])
    product_mrp = st.slider("Product MRP (price)", 31.0, 266.0, 147.0, 0.5)

with col2:
    # Store Features
    st.markdown("**Store Details**")
    establishment_year = st.slider("Store Establishment Year",
                                1987, datetime.now().year, 2002)
    store_size = st.selectbox("Store Size", ["Small", "Medium", "Large"])
    city_type = st.selectbox("City Tier", ["Tier 1", "Tier 2", "Tier 3"])
    store_type = st.selectbox("Store Type", [
        "Supermarket Type1", "Supermarket Type2",
        "Departmental Store", "Food Mart"
    ])

# Prepare input data (only including features used in the backend)
input_data = {
    "product_weight": product_weight,
    "product_allocated_area": product_area,
    "product_mrp": product_mrp,
    "store_establishment_year": establishment_year,
    "product_sugar_content": product_sugar,
    "product_type": product_type,
    "store_size": store_size,
    "store_location_city_type": city_type,
    "store_type": store_type
}

# Make prediction when the "Predict" button is clicked
if st.button("Predict Sales"):
    try:
        # Updated API endpoint
        backend_url = "https://DD009-SuperKartBackend.hf.space"  # Corrected URL
        endpoint = f"{backend_url}/v1/sales"
        
        st.info(f"Connecting to: {endpoint}")  # Show the endpoint being called
        
        response = requests.post(endpoint, json=input_data, timeout=10)
        
        if response.status_code == 200:
            try:
                result = response.json()
                st.success(f"Predicted Sales Total: ${result['predicted_sales']:.2f}")
                
                # Display features used
                st.markdown("**Features Used**")
                st.write(", ".join(result['features_used']))
            except ValueError:
                st.error("Could not decode JSON response from server")
                st.text(f"Raw response: {response.text}")
        else:
            st.error(f"Error making prediction (Status {response.status_code})")
            try:
                error_details = response.json()
                st.json(error_details)
            except ValueError:
                st.text(f"Raw response: {response.text}")
    except requests.exceptions.RequestException as e:
        st.error(f"Connection error: {str(e)}")
        st.info("Please check if the backend server is running and accessible")

# Section for batch prediction
st.subheader("Batch Prediction")
st.write("Upload a CSV file containing multiple product-store combinations")

# Instructions for CSV format
st.markdown("""
**CSV File Requirements:**
- Must contain these exact columns:
  - Product_Weight, Product_Allocated_Area, Product_MRP
  - Store_Establishment_Year, Product_Sugar_Content, Product_Type
  - Store_Size, Store_Location_City_Type, Store_Type
""")

# File uploader for batch predictions
uploaded_file = st.file_uploader("Choose a CSV file", type="csv")

if uploaded_file is not None:
    if st.button("Predict Batch Sales"):
        try:
            backend_url = "https://DD009-SuperKartBackend.hf.space"  # Corrected URL
            endpoint = f"{backend_url}/v1/salesbatch"
            
            st.info(f"Connecting to: {endpoint}")  # Show the endpoint being called
            
            files = {'file': (uploaded_file.name, uploaded_file, 'text/csv')}
            response = requests.post(endpoint, files=files, timeout=30)

            if response.status_code == 200:
                try:
                    results = response.json()
                    if 'predictions' in results:
                        results_df = pd.DataFrame(results['predictions'])
                        st.success("Batch predictions completed!")
                        st.dataframe(results_df)

                        # Download button for results
                        csv = results_df.to_csv(index=False)
                        st.download_button(
                            label="Download predictions as CSV",
                            data=csv,
                            file_name='sales_predictions.csv',
                            mime='text/csv'
                        )
                    else:
                        st.error("Unexpected response format from server")
                        st.json(results)
                except ValueError:
                    st.error("Could not decode JSON response from server")
                    st.text(f"Raw response: {response.text}")
            else:
                st.error(f"Error making predictions (Status {response.status_code})")
                try:
                    error_details = response.json()
                    st.json(error_details)
                except ValueError:
                    st.text(f"Raw response: {response.text}")
        except requests.exceptions.RequestException as e:
            st.error(f"Connection error: {str(e)}")
            st.info("Please check if the backend server is running and accessible")

# Add sample data section
st.sidebar.markdown("### Sample Data")
if st.sidebar.button("Show Sample Input"):
    sample_data = {
        "product_weight": 12.66,
        "product_allocated_area": 0.027,
        "product_mrp": 117.08,
        "store_establishment_year": 2009,
        "product_sugar_content": "Low",
        "product_type": "Frozen Foods",
        "store_size": "Medium",
        "store_location_city_type": "Tier 2",
        "store_type": "Supermarket Type2"
    }
    
    st.sidebar.json(sample_data)
    st.sidebar.download_button(
        label="Download Sample CSV",
        data=pd.DataFrame([sample_data]).to_csv(index=False),
        file_name='sample_input.csv',
        mime='text/csv'
    )