Welcome to the PowerCast: Forecasting Electricity Demand from Weather Conditions repository! ⚡️
This project is a collaborative initiative brought to you by SuperDataScience, a global learning community focused on data science, machine learning, and AI. Whether you're just getting started or diving deeper into sequence modeling, we're excited to have you here!
To contribute to this project, please follow the steps outlined in our CONTRIBUTING.md file.
This project supports two tracks based on experience level:
SDS-CP036-powercast/
├── beginner/ ← Beginner track files
│ ├── README.md ← Scope of Works for Beginner Track
│ ├── REPORT.md ← Markdown template for beginner submissions
│ └── submissions/
│ ├── team-members/
│ └── community-contributions/
│
├── advanced/ ← Advanced track files
│ ├── README.md ← Scope of Works for Advanced Track
│ ├── REPORT.md ← Markdown template for advanced submissions
│ └── submissions/
│ ├── team-members/
│ └── community-contributions/
│
├── CONTRIBUTING.md
├── requirements.txt
└── README.md ← You are here!
The Beginner Track is great for learners looking to build regression models from tabular environmental time-series data. You’ll analyze how weather features like temperature, humidity, and solar radiation impact power consumption in Tetouan City. You'll use models like Linear Regression, Random Forest, and XGBoost, and deploy your final model with Streamlit.
📌 Get started: ➡️ Beginner Track Scope of Works ➡️ Beginner Report Template ➡️ Submit your work
The Advanced Track is for those ready to explore sequence modeling using LSTM, GRU, or Temporal Convolutional Networks. You’ll design time-aware neural networks, experiment with lag features and sliding windows, and optionally include model explainability tools. Deployment may be done via Docker, Hugging Face Spaces, or APIs.
📌 Get started: ➡️ Advanced Track Scope of Works ➡️ Advanced Report Template ➡️ Submit your work
We’ll be working with real-time weather and electricity consumption data:
🔗 Tetouan City Power Consumption Dataset
| Phase | General Activities |
|---|---|
| Week 1: Setup + EDA | Clean and inspect time-series data, uncover patterns and relationships |
| Week 2: Feature Engineering | Lag features, rolling stats, encode time, normalize data |
| Week 3: Model Development | Train regression models and compare zone-wise performance |
| Week 4: Model Optimization | Tune hyperparameters and finalize your best models |
| Week 5: Deployment | Build and host your Streamlit power forecasting app |
This project is open to both official team members and outside community contributors.
- 🧑💻 Team Members should submit their work under
team-members/ - 🌍 Community Contributors are welcome to fork the repo and submit under
community-contributions/
See CONTRIBUTING.md for full participation guidelines.