⚡ Electricity_Cost_Prediction_API

Welcome to the Electricity_Cost_Prediction_API. This project will predict monthly electricity costs for residential and commercial structures using GradientBoostingRegressor Machie Learning algorithm. this API is designed to provide accurate predictions based on environmental and structural features.

---

🌐 Live API

Base URL: https://electricity-cost-api-m893.onrender.com/docs

---

📌 Endpoints

. GET /

Welcome Page.

Response:

{"message":"Welcome to the electricity cost  prediction API!"}

. GET /health

This endpoint will assure you that your api is working fine and your model is loading properly.

Response:

{
  "status": "ok",
  "model_loaded": true
}

. POST /prediction

Submit the data according to model and get the output.

Request Body:

{
  "site_area": 100,
  "structure_type": "Mixed-use",
  "water_consumption": 10,
  "recycling_rate": 20,
  "utilisation_rate": 10,
  "air_quality_index": 80,
  "issue_resolution_time": 10,
  "resident_count": 19
}

Response:

{
  "electricity_cost": 6130
}

---

▀▄▀▄▀▄ PROJECT OVERVIEW ▄▀▄▀▄▀

This project is more than just an API; it's a complete pipeline that includes:

. Exploratory Data Analysis (EDA) and preprocessing of CSV data
. Training and evaluation of a Machine Learning regression model
. FastAPI backend for seamless predictions
. Live deployment on Render for global accessibility

---

🧠 Model Features

The prediction model considers these key parameters:

Feature Category	Description	Units/Values
Geospatial
Site Area	Total area of the site	Square meters (m²)
Structure Type	Building classification	Residential/Commercial/Mixed-use
Resource Usage
Water Consumption	Monthly water usage	Liters (L)
Recycling Rate	Percentage of waste recycled	Percentage (%)
Utilisation Rate	Percentage of resource utilization	Percentage (%)
Environmental
Air Quality Index	Measure of air pollution (higher = worse air quality)	0-500 scale
Operational
Issue Resolution Time	Average time to resolve maintenance issues	Hours (h)
Resident Count	Number of regular occupants	Count

---

📈 Model Performance

📊 Evaluation Metrics

Metric	Score	Interpretation
R² Score	0.96	Explains 96% of cost variance
MSE	46,886.6	Mean squared error
MAE	173.38	Mean absolute error in cost units

---

☁ Deployment on Render

The API is deployed on Render.

✅ FastAPI app served with Uvicorn
✅ Automated deployment using docker image(https://hub.docker.com/repository/docker/aryan020/electricity-cost-api5/tags)

✅ Accessible at: https://electricity-cost-api-m893.onrender.com/docs

---

📁 Project Structure

ElectricityCostPrediction/
│
├── model/
│   ├── electricity_cost_dataset.csv  # Dataset used for training
│   └── model.joblib                 # Trained ML model (serialized)
│
├── schema/
│   └── user_input.py                # Pydantic models for data validation
│
├── app.py                           # FastAPI application entry point
├── Dockerfile                       # Containerization configuration
├── requirements.txt                 # Python dependencies
├── README.md                        # Project documentation
├── frontend.py                      # Optional frontend interface
└── electricity.ipynb                # Jupyter notebook (EDA/feature engineering/model training)

---

🚀 Local Setup

Want to run api locally? Follow these steps 1.Clone the repository and go to main folder by cd command.

git clone https://github.com/Dibbu-cell/Electricity_Cost_Prediction.git
cd Electricity_Cost_Prediction

2.Create Virtual Envoirenment and activate it .

python -m venv venv
venv\Scripts\activate   #for window users

3.Install Dependcies

pip install -r requirements.txt

4. Run api locally

uvicorn app:app --reload

5.Open your browser and navigate to: http://127.0.0.1:8000/docs

---

📌 Frontend setup

If you want to experience through frontend do these step.

1. Firstly do local set up mention above after that run these commands .
2. If you are ruuning the api in your machine then change API_LINK to (http://127.0.0.1:8000/prediction) in your( frontend.py) file .
3. If you are running the api using this link (https://electricity-cost-api-m893.onrender.com/docs) then no need to change API_LINK.
4. install streamlit in your envoirnment.

pip install streamlit

5.run frontend.py and get intractive user interface.

 streamlit run frontend.py

6. You will get intractive user interface and you can update the values and predict the electricity cost.

---

I hope you find this electricity cost prediction api good enough . If you have any doubt feel free to ask me through mail ** [email protected]**