Chan-dre-yi/POWER-CAST

This project is a time series forecasting model using the Temporal Fusion Transformer (TFT) deep learning architecture. The model is trained and evaluated on the M4 competition dataset, achieving state-of-the-art results in multi-step forecasting tasks.

22
/ 100
Experimental

This project helps operations managers and energy analysts predict future energy demand for utilities and grids. You provide historical energy consumption data, and it outputs multi-step forecasts showing expected demand. It's designed for professionals who need accurate predictions to manage resources and plan operations.

No commits in the last 6 months.

Use this if you need highly accurate, multi-step predictions for time-sensitive data like energy consumption, especially when long-term patterns and seasonal variations are important.

Not ideal if you're dealing with very short, simple time series or if you prefer simpler, more traditional statistical forecasting methods.

energy-forecasting demand-planning utility-operations grid-management resource-allocation
No License Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 4 / 25
Maturity 8 / 25
Community 8 / 25

How are scores calculated?

Stars

8

Forks

1

Language

Jupyter Notebook

License

Last pushed

Jul 17, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Chan-dre-yi/POWER-CAST"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.