algorithmic-efficiency icon indicating copy to clipboard operation
algorithmic-efficiency copied to clipboard

Add workload variants

Open priyakasimbeg opened this issue 2 years ago • 2 comments

Add workload variants for the base workloads.

This is a tracking issue. 7/8 variants along with model-diff tests have been added already.

Remaining work is to:

  • [x] Submit DeepSpeech workloads without OOMs.
  • [ ] Run all variants e2e.

priyakasimbeg avatar Jan 17 '24 02:01 priyakasimbeg

Remaining tasks.

Fix

  • [x] https://github.com/mlcommons/algorithmic-efficiency/issues/654
  • [x] https://github.com/mlcommons/algorithmic-efficiency/issues/662
  • [x] Criteo embed init scale workload repository variant error

Rerun variants for:

  • [x] criteo embed init scale
  • [x] criteo resnet
  • [x] imagenet_resnet_silu
  • [x] imagenet_resnet_gelu
  • [x] vit GLU
  • [x] conformer gelu
  • [x] conformer layernorm
  • [x] conformer attention temp
  • [x] deepspeech specaug
  • [x] wmt attention temp

priyakasimbeg avatar Feb 29 '24 22:02 priyakasimbeg

Pending variant reruns with correct hyperparamters for:

  • [x] deepspeech_norm_and_spec_aug
  • [x] deepspeech_tanh
  • [x] deepspeech_no_resnet

priyakasimbeg avatar Mar 29 '24 00:03 priyakasimbeg