Skip to content

Conversation

runame
Copy link
Contributor

@runame runame commented Aug 25, 2022

This PR adds the current configurations of the target-setting runs, including Jax and PyTorch implementations of the AdamW, NAdamW, and Nesterov optimizers.

Still missing:

  • Interface for label smoothing and dropout which is accessible to the user (currently affects the ImageNet-ResNet and the WMT target-setting run),
  • target-setting runs for workloads which are not fully implemented in this repo (PyTorch Criteo1TB DLRM-Small, Jax FastMRI + all Librispeech workloads).
@runame runame requested a review from znado August 25, 2022 17:59
@github-actions
Copy link

github-actions bot commented Aug 25, 2022

MLCommons CLA bot All contributors have signed the MLCommons CLA ✍️ ✅

@znado znado merged commit f3c5d86 into mlcommons:main Aug 31, 2022
@github-actions github-actions bot locked and limited conversation to collaborators Aug 31, 2022
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
3 participants