Home / Platform / Forecaster
Platform Layer: Modular toolkit active time-series-forecasting

Forecaster

Forecaster is a web-based platform for time-series forecasting. It streamlines training, evaluating, and comparing forecasting approaches, for users without deep technical expertise.

Features

  • Built-in tools for missing data and preprocessing
  • Language-model recommendations for parameter selection and result interpretation
  • Intuitive visualizations
  • Multiple forecasting models: linear regression, N-Linear, and the Temporal Fusion Transformer
  • A web interface that requires no programming

Part of the Kentucky Open Science platform: it runs over the Cresco mesh, with versioned storage for datasets, models, and code in GaiaKeep. See the full stack.