Or put another way , how well does it scale horizontally to multiple machines.
We fine most time series libraries to be about the same in terms of features and speed, but very few can handle large datasets well, if at all.
And, of course, thanks for sharing your library, I'll definitely try it out!!
The models that work on multiple time series in Darts accept Sequence[TimeSeries] for their fit() method. These sequences can either be Lists (fully in memory, simplest option), or when needed it can be a custom Sequence which for example does lazy loading from disk (somewhat similar to what PyTorch Datasets are doing) with the __getitem__() method.
If you need even more control, for instance because you have only one very long series that doesn't fit in memory then you can implement your own Darts "TrainingDataset". In this case you can control how to slice your series exactly.
Edit: I realised this only answers the first sentence of your comment ;) For now there's no mechanism for scaling to multiple machines beyond what PyTorch is already offering. AFAIK it's reasonably easy to scale to multiple GPUs on a machine, but I'm not sure how it would scale on several machines. We never had to try this yet! (Note that actually a single CPU can handle training deep nets models on 10's of thousands of time series similar to the M4 competition in a fairly reasonable time).
I'd be curious about the performance of these. A time series featurization library I've liked the look of but haven't used for real is catch22: https://github.com/chlubba/catch22
In particular I like catch22's methodology:
catch22 is a collection of 22 time-series [that are] are a high-performing subset of the over 7000 features in hctsa. Features were selected based on their classification performance across a collection of 93 real-world time-series classification problems...
This series might be useful to you. https://www.youtube.com/watch?v=ZoJ2OctrFLA&list=PLvcbYUQ5t0...
It seems that almost everywhere you look, every example has just one timeseries that needs to be dealt with. However, since the methods are much more "statistical" in nature, they can actually make meaningful predictions on a single sample.
These automatic extractions are very statistical in nature indeed, but for some datasets domain insights are more valuable and give more usable features (in my opinion). I found quite some datasets where manual features + gradient boosted trees give better results then automated statistical methods. Often combinations give better results :)
[1]: https://github.com/unit8co/darts/
[2]: https://medium.com/unit8-machine-learning-publication/traini...
I worked with this algorithm before so I was curious, but I can't find it in the API.
