app.py

import streamlit as st
import pandas as pd
import requests

# =======================================================
# SuperKart Price Prediction (Streamlit Frontend)
# =======================================================

# Title of the Streamlit web app
st.title("🛒 SuperKart Price Prediction")

# -------------------------------------------------------
# Section 1: Online (Single) Prediction
# -------------------------------------------------------
st.subheader("🔮 Predict Price for a Single Product")

# Collect user input for product features
product_name = st.text_input("Product Name", "Sample Product")  # Text input field for product name
brand = st.text_input("Brand", "BrandX")                       # Text input for brand name
category = st.selectbox("Category", ["Electronics", "Clothing", "Groceries", "Home", "Other"])  
# Dropdown menu to select product category

ratings = st.number_input("Customer Ratings (0 - 5)", min_value=0.0, max_value=5.0, value=4.0, step=0.1)  
# Numeric input for product rating

discount = st.number_input("Discount (%)", min_value=0.0, max_value=100.0, value=10.0, step=0.5)  
# Numeric input for discount percentage

# Create a DataFrame to format inputs for the backend API
input_data = pd.DataFrame([{
    "product_name": product_name,
    "brand": brand,
    "category": category,
    "ratings": ratings,
    "discount": discount
}])

# When the "Predict Price" button is clicked → send request to backend
if st.button("Predict Price"):
    try:
        # Send JSON payload to the Flask backend API hosted on Hugging Face
        response = requests.post(
            "https://JefferyMendis-SuperKartBackend.hf.space/v1/predict",
            json=input_data.to_dict(orient="records")[0]
        )

        # If successful → display prediction
        if response.status_code == 200:
            prediction = response.json()["Predicted Price (in dollars)"]
            st.success(f"💰 Predicted Selling Price: ${prediction:.2f}")
        else:
            st.error("⚠️ Error making prediction. Please check backend logs.")
    except Exception as e:
        # Handle connection issues gracefully
        st.error(f"⚠️ Connection error: {e}")

# -------------------------------------------------------
# Section 2: Batch Prediction (CSV Upload)
# -------------------------------------------------------
st.subheader("📂 Batch Prediction (Upload CSV)")

# File uploader widget for batch prediction
uploaded_file = st.file_uploader("Upload CSV file with product data", type=["csv"])

# When user uploads a file and clicks button → send file to backend
if uploaded_file is not None:
    if st.button("Predict Batch"):
        try:
            # Send file to backend API
            response = requests.post(
                "https://JefferyMendis-SuperKartBackend.hf.space/v1/predictbatch",
                files={"file": uploaded_file}
            )

            # If successful → display results
            if response.status_code == 200:
                predictions = response.json()
                st.success("✅ Batch predictions completed!")
                st.write(predictions)  # Display predictions in table format
            else:
                st.error("⚠️ Error in batch prediction. Check backend.")
        except Exception as e:
            st.error(f"⚠️ Connection error: {e}")