Welcome to spotforecast2


Version GitHub Python

spotforecast2 is a Python package for forecasting, combining the power of sklearn, spotoptim and skforecast with specialized utilities for “spot” forecasting.

Installation

  • Download from GitHub
git clone https://github.com/sequential-parameter-optimization/spotforecast2.git
cd spotforecast2
  • Sync using uv
uv sync

Features

  • Data Fetching: Easy access to time series data.
  • Preprocessing: Robust tools for curating, cleaning, and splitting data.
  • Forecasting: A rich set of forecasting strategies (constantly extended).
  • Model Selection: spotoptim and optuna search for hyperparameter tuning.
  • Weather Integration: Utilities for fetching and using weather data in forecasts.

Attributions

Parts of the code are ported from skforecast to reduce external dependencies. Many thanks to the skforecast team for their great work!