ThomasBourgeois

Results 9 comments of ThomasBourgeois

That would clearly be a great add-on.

Temporarilly had to go to fastai 2.6.3, and seems to work

Hi @AxeldeRomblay , thx for your reply, I hadnt seen this was already referenced. Wow, tweaking the code with throw-away lines, that is a no-go for me, but if you...

nope I meant [there](https://github.com/Nixtla/statsforecast/blob/main/examples/arima.ipynb) I can take care of it by the way if needed, I've never pushed to github, that would be a first contribution.

> An ARIMA model with zero means is white noise, so it means that the errors are uncorrelated across time. > > This doesn't imply anything about the size of...

Thanks ! Golden features were still unclear to me, I wondered if you did one big model with all new features created and used features_importances. So I checked the code...

Thanks for your answer ! The example is quite straightforward, thanks. Maybe one comment : 'cv' is sometimes used in classic libraries such as scikit learn as short for 'cross...

Ok get it, I thought it meant custom validation, I understand now you can pass multiple tuples and do a cross validation, hence why you call it cv !

Interested, even if there needs to be some coding involved too.