Stability of the API

[hrzn]

Hi, great work on statsforecasts! This package looks very nice.
I'm considering integrating a couple of the models in Darts (https://github.com/unit8co/darts). I'm wondering about your future plans - do you intend to maintain this package on the long term? How likely can we expect API changes in the future releases?

Also as a side note - I took a quick look at the Croston method, and it looks like the method accepts h and future_xreg, which I'm not sure is intended as those are not used.

In general I think slightly more documentations on your different models could be helpful for users :)

[mergenthaler]

Hi! Thank you very much. We are happy about the integration with Darts, we really like your library. We intend to give full maintenance and support to this package and we plan for the APIs to be as stable as possible. Thanks for the croston observation. And thanks for the feedback on documentation.

[hrzn]

Perfect, thanks!
fyi: here's the PR into Darts to add AutoARIMA (called AutoARIMASF for the time being) and Croston: unit8co/darts#893.
Is there anything else you'd recommend we should take a look at?

[AzulGarza]

Hi @hrzn ! It would be great if the class name was StatsForecastAutoARIMA to match other libraries (sktime for example). Thanks! :)

[hrzn]

Hi @hrzn ! It would be great if the class name was StatsForecastAutoARIMA to match other libraries (sktime for example). Thanks! :)

Yes sure, I can do that 👍

[AzulGarza]

Hi @hrzn. I've seen that StatsForecast is now included in darts. Thank you! Nice work!
Let us know if you have any more questions or feature requests.

[hrzn]

Thanks for your work @FedericoGarza. Indeed it was released very recently and I'm very happy to see it in Darts. I'll let you know if I have any feedback (please do the same).