Stas Bekman

@empty-id, make sure you have cuda-11.1 or higher installed and configured correctly - please see: https://huggingface.co/transformers/master/main_classes/trainer.html#possible-problem-2

Looks like the same error as reported here https://github.com/pytorch/pytorch/issues/47669#issuecomment-725073808 which apparently has been fixed in pytorch many months back. Try pytorch-1.9.0 and if it doesn't work please file a new...

His is Command line client (version 1.6.14) But I can't see how that information would help me, if the client is not behaving properly. Surely it should have said that...

I did some more testing and was able to create (via `conda skeleton pypi`), build and upload other modules w/o any problems. I also tested that this package fails with...

it won't let me upload this type of file, here it is on [dropbox](https://www.dropbox.com/s/sdtqsu72barm81d/fastprogress-0.1.5-py36_0.tar.bz2?dl=0) And thank you for your help, @goanpeca

Were you able to get the package via dropbox, @goanpeca?

some additional info: * I'm able to upload this package under a different username `-u fastai`, a group, to which I'm allowed to upload to. * I'm able to upload...

@thomasw21, please feel free to close this one or build on top of it, either way works. I think we should test too that `original_pp_degree != 0`

yes, it will always be larger as the training progresses, yet we need to continue skipping earlier ranges. so with `--skip-train-iteration-range 13251-14000 16651-19500` the actual iteration will typically be somewhere...

> > But also say we want to re-train the whole thing keeping these ranges, then iteration can be anything - smaller and larger. > > This totally makes sense....