Guys, I've lost some of my result, because couldn't reproduce it . Is it any "magic" seeding formula, that guaranty repeating results, that i can use at the beginning of my notebook?
I think we need to know first the kind of model framework you used. Is it pytorch, fastai or tensorflow?
in this case I've used fastai
Okay. Check this link out:
fastai2 reproducibility · Issue #2832 · fastai/fastai (github.com)
Hope it helps!!!!
Thanks! i'll try it in next competition :)
I think we need to know first the kind of model framework you used. Is it pytorch, fastai or tensorflow?
in this case I've used fastai
Okay. Check this link out:
fastai2 reproducibility · Issue #2832 · fastai/fastai (github.com)
Hope it helps!!!!
Thanks! i'll try it in next competition :)