Mattis Megevand
Mattis Megevand
Done
Forgot to set the seed for tensors, should be good now
There’s no reason in particular both optimizers works with the schedulers. Since for the testing we're training on MNIST it was easy for the model to reach convergence without a...
Closing this PR as this isn't problematic and avoiding breaking changes especially since customers are not complaining about this discrepancy between the way we create the description in sale and...
Hey, I was working on this a bit earlier before I saw that you had a PR for this bounty. I think we should cache necessary reshape and permutation given...