Darts (https://github.com/unit8co/darts/) is a library making it very easy to use a wide range of forecasting models, from classic statistical models (such as ARIMA, Exponential smoothing), to recent state-of-the-art deep learning models (such as N-BEATS or TCN). All models use a simple fit/predict interface similar to sklearn, which makes it very easy to switch and compare models. The library also has other features such as easy backtesting, ensembling, and model selection, among others. Finally, the deep learning models can be trained on multiple time series and also support multivariate series.
We would be very happy to see added to the list of awesome libs, along other relevant related projects :)
Darts (https://github.com/unit8co/darts/) is a library making it very easy to use a wide range of forecasting models, from classic statistical models (such as ARIMA, Exponential smoothing), to recent state-of-the-art deep learning models (such as N-BEATS or TCN). All models use a simple fit/predict interface similar to sklearn, which makes it very easy to switch and compare models. The library also has other features such as easy backtesting, ensembling, and model selection, among others. Finally, the deep learning models can be trained on multiple time series and also support multivariate series.
We would be very happy to see added to the list of awesome libs, along other relevant related projects :)
Many thanks, Julien