Get the latest tech news
The explosion in time series forecasting packages in data science
Arthur Turrell is an economic data scientist.
It’s worth saying that while these new packages bring lots that is good, like state of the art (SOTA) machine learning (ML) models, and super modern fast fitting algorithms, their understandable focus on the needs of tech firms can mean they’re less well-suited, at least out of the box, to lower frequency data. A lightweight, easy-to-use, generalizable, and extendable framework to perform time series analysis, from understanding the key statistics and characteristics, detecting change points and anomalies, to forecasting future trends. tsai is an open-source deep learning package built on top of Pytorch & fastai focused on state-of-the-art techniques for time series tasks like classification, regression, forecasting, imputation.
Or read this on Hacker News
