Update README.md

README.md

@@ -16,6 +16,99 @@ The backend is designed to be consumed by the **frontend Streamlit app** (`mwill
 
 ---
 
-## 🚀 Live API Endpoints
+## Live API Endpoints
+
+Base URL:  https://mwill-aimission-sales-forecast-model.hf.space/
+
+
+### 1. Root – Welcome
+`GET /`  
+Returns a welcome message confirming the API is live.
+
+Example:
+```bash
+curl https://mwill-aimission-sales-forecast-model.hf.space/
+
+Response:
+"Welcome to the SuperKart Sales Forecasting API!"
+
+
+### 2. Health Check
+GET /health
+Returns a simple status object to confirm the service is running.
+
+Example:
+curl https://mwill-aimission-sales-forecast-model.hf.space/health
+Response:
+{"status": "ok"}
+
+
+### 3. Endpoint List
+GET /endpoints
+Lists all available API endpoints and their descriptions.
+Example:
+curl https://mwill-aimission-sales-forecast-model.hf.space/endpoints
+
+Response:
+{
+  "GET /": "Welcome page",
+  "GET /health": "Simple status check",
+  "POST /v1/predict": "Send JSON; returns predicted_sales"
+}
+
+### 4. Predict Sales
+POST /v1/predict
+Accepts a JSON payload containing product and store attributes, calculates engineered features, preprocesses the input, and returns the predicted sales total.
+
+Expected JSON Payload:
+{
+  "Product_Weight": 12.0,
+  "Product_Sugar_Content": "Low Sugar",
+  "Product_Allocated_Area": 0.068,
+  "Product_MRP": 147.03,
+  "Store_Size": "Medium",
+  "Store_Location_City_Type": "Tier 2",
+  "Store_Type": "Supermarket Type2",
+  "Product_Type": "Snack Foods",
+  "Store_Id": "OUT003",
+  "Store_Establishment_Year": 1998
+}
+
+Example Request:
+curl -X POST \
+  https://mwill-aimission-sales-forecast-model.hf.space/v1/predict \
+  -H "Content-Type: application/json" \
+  -d '{
+        "Product_Weight": 12.0,
+        "Product_Sugar_Content": "Low Sugar",
+        "Product_Allocated_Area": 0.068,
+        "Product_MRP": 147.03,
+        "Store_Size": "Medium",
+        "Store_Location_City_Type": "Tier 2",
+        "Store_Type": "Supermarket Type2",
+        "Product_Type": "Snack Foods",
+        "Store_Id": "OUT003",
+        "Store_Establishment_Year": 1998
+      }'
+
+Example Response:
+{
+  "predicted_sales": 7614.54
+}
+
+🛠️ Tech Stack
+Language: Python 3.10+
+Framework: Flask
+ML Libraries: scikit-learn, pandas, numpy
+Deployment: Hugging Face Spaces (Docker)
+
+📌 Notes
+The model is preloaded at startup for performance.
+Feature engineering (Store_Age, Product_Allocated_Area_Sq, and Revenue_per_Allocated_Area) happens in the backend before prediction.
+The API is stateless and ready to be integrated with any frontend or automation pipeline.
+
+🔗 Related Project
+Frontend Streamlit App:
+https://huggingface.co/spaces/mwill-AImission/superkart-sales-forecast-frontend
+
 
-Base URL:  https://mwill-aimission-sales-forecast-model.hf.space/