metaforecast 0.1.3

Meta-learning and Data-centric Forecasting

metaforecast: Meta-Learning and Data-Centric AI for Actionable Forecasting

metaforecast is a Python package that combines meta-learning techniques with data-centric AI approaches to provide powerful and actionable forecasting capabilities. Built on top of the Nixtla ecosystem, this package offers advanced tools for time series analysis and prediction.

Features

MetaForecast currently consists of three main modules:

1. Dynamic Ensembles: Leveraging multiple models with adaptive ensemble techniques.
2. Synthetic Time Series Generation: Creating realistic synthetic time series data for robust model training and testing.
3. Long-Horizon Meta-Learning: Instance-based meta-learning for multi-step forecasting.

Installation

You can install metaforecast using pip:

pip install metaforecast

Quick Start

todo

Documentation

For detailed documentation, please visit todo

Examples

Check out the notebooks folder (currently under construction) for example usage and tutorials.

Dependencies

metaforecast is built on top of the Nixtla ecosystem.