Update app.py

app.py

@@ -1,39 +1,115 @@
-
-from flask import Flask, request, jsonify
-import joblib
+import os
+import streamlit as st
+import requests
 import pandas as pd
 
-app = Flask(__name__)
+st.set_page_config(page_title="🛒 SuperKart Sales Forecast", layout="centered")
+
+st.title("SuperKart Sales Forecast")
+st.caption("Frontend powered by Streamlit → calls Flask backend for predictions")
+
+# -----------------------------
+# Backend URL Resolver
+# -----------------------------
+def resolve_backend_url() -> str:
+    # Prefer secret, then env, then hardcoded fallback
+    url = None
+    try:
+        url = st.secrets.get("BACKEND_URL")
+    except Exception:
+        pass
+    url = url or os.getenv("BACKEND_URL") or "https://rizwan9-backend.hf.space"
+    return url.strip()
 
-# Load the serialized model
-model = joblib.load('best_sales_forecasting_model.pkl')
+BACKEND_URL = resolve_backend_url()
 
-@app.route('/')
-def home():
-    return "Sales Forecasting Backend is running!"
+# -----------------------------
+# Sidebar: Backend Health
+# -----------------------------
+st.sidebar.title("⚙️ Backend")
+backend_url_input = st.sidebar.text_input("Backend URL", value=BACKEND_URL)
+BACKEND_URL = backend_url_input.strip() or BACKEND_URL
+st.sidebar.markdown(f"**URL:** [{BACKEND_URL}]({BACKEND_URL})")
 
-@app.route('/predict', methods=['POST'])
-def predict():
+@st.cache_data(ttl=60, show_spinner=False)
+def check_backend_health(url: str, timeout: int = 45):
     try:
-        data = request.get_json(force=True)
-        # Convert the incoming data to a pandas DataFrame
-        # Assuming the incoming data is a list of dictionaries, where each dictionary is a data point
-        input_data = pd.DataFrame(data)
-
-        # Ensure the columns are in the same order as the training data
-        # This assumes you have access to the columns from your training data (X_train)
-        # You might need to adjust this part based on how you handle feature ordering
-        # For demonstration, let's assume the input data has the same columns in the same order
-        # In a real application, you might need to reorder columns or handle missing ones
-
-        # Make predictions
-        predictions = model.predict(input_data)
-
-        # Return predictions as a JSON response
-        return jsonify(predictions.tolist())
+        r = requests.get(f"{url}/health", timeout=timeout)
+        return r.status_code, r.text
     except Exception as e:
-        return jsonify({'error': str(e)})
+        return None, str(e)
+
+status_box = st.sidebar.empty()
+auto_check = st.sidebar.toggle("Auto check health on load", value=True)
+
+def run_health_check():
+    with status_box, st.spinner("Checking backend health..."):
+        code, msg = check_backend_health(BACKEND_URL)
+    if code == 200:
+        status_box.success("✅ Healthy (200)")
+    elif code is None:
+        status_box.error(f"❌ Unreachable\n{msg}")
+    else:
+        status_box.warning(f"⚠️ Status {code}\n{msg}")
+
+if auto_check:
+    run_health_check()
+
+if st.sidebar.button("🔍 Check Health Now"):
+    check_backend_health.clear()   # clear cache
+    run_health_check()
 
-if __name__ == '__main__':
-    # Running on 0.0.0.0 makes it accessible externally, useful for deployment
-    app.run(host='0.0.0.0', port=5000)
+st.divider()
+st.subheader("Enter Product and Store Details")
+
+# -----------------------------
+# Main Form
+# -----------------------------
+with st.form("input_form"):
+    col1, col2 = st.columns(2)
+
+    with col1:
+        Product_Weight = st.number_input("Product Weight", min_value=0.0, step=0.1)
+        Product_Allocated_Area = st.number_input("Product Allocated Area", min_value=0.0, step=0.001, format="%.3f")
+        Product_MRP = st.number_input("Product MRP", min_value=0.0, step=0.5)
+        Store_Establishment_Year = st.number_input("Store Establishment Year", min_value=1950, max_value=2025, step=1)
+
+    with col2:
+        Product_Sugar_Content = st.selectbox("Product Sugar Content", ["Low Sugar", "Regular", "No Sugar"])
+        # Use dataset's exact labels (note lowercase 'and')
+        Product_Type = st.selectbox("Product Type", [
+            "Meat","Snack Foods","Hard Drinks","Dairy","Canned","Soft Drinks","Health and Hygiene",
+            "Baking Goods","Bread","Breakfast","Frozen Foods","Fruits and Vegetables","Household",
+            "Seafood","Starchy Foods","Others"
+        ])
+        Store_Size = st.selectbox("Store Size", ["Low","Medium","High"])
+        Store_Location_City_Type = st.selectbox("City Tier", ["Tier 1","Tier 2","Tier 3"])
+        Store_Type = st.selectbox("Store Type", ["Departmental Store","Supermarket Type 1","Supermarket Type 2","Food Mart"])
+
+    submitted = st.form_submit_button("🚀 Predict Sales")
+
+# -----------------------------
+# Call backend
+# -----------------------------
+if submitted:
+    payload = {
+        "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_Establishment_Year": int(Store_Establishment_Year),
+        "Store_Size": Store_Size,
+        "Store_Location_City_Type": Store_Location_City_Type,
+        "Store_Type": Store_Type
+    }
+
+    try:
+        with st.spinner("Fetching prediction from backend..."):
+            r = requests.post(f"{BACKEND_URL}/predict", json=payload, timeout=60)
+            r.raise_for_status()
+            data = r.json()
+            prediction = data["predictions"][0]
+        st.success(f"📈 Predicted Product Store Sales Total: **{prediction:,.2f}**")
+    except Exception as e:
+        st.error(f"❌ Prediction failed:\n\n{e}")