Collaboration

[microprediction]

Hi this is great. Can't believe I didn't notice your repo.

I am a collector of univariate TS methods. I would like to massage some of your stuff into the "skater" format I use (see skater signature) so as to include it downstream in all sorts of things (see https://github.com/microprediction for explanation).

The reverse trip looks pretty easy ... just looking at you https://github.com/ITMO-NSS-team/pytsbe/blob/main/pytsbe/models/pmdarima_forecaster.py now.

Btw I can provide lots of fresh data if you want it. See https://microprediction.github.io/microprediction/data.html and help yourself.

-peter

[Dreamlone]

Hi! Thank you very much for this issue, I am glad to hear that you liked the repository :)

Sure, feel free to use and adapt the methods that are presented in this library and in the FEDOT (Automl tool for regression, classification and time series forecasting tasks). Or integrate your algorithms into pytsbe also. I am not currently an owner of this repository, nor of the FEDOT framework. So it might be better to contact with Nikolay if you need help with adaptation. Or if there are some ideas you want to discuss. Nevertheless, I'll try to participate too, if I have time and energy to do that (after my regular job tasks) :)

About the data - thank you very much, perhaps it could indeed be added to the pytsbe. If you would like to test the algorithms on your data with pytsbe, I would be glad to help you!

Sorry for the late reply

[microprediction]

I'll give it a spin, thx for the reply.