issues
search
mlcommons
/
algorithmic-efficiency
MLCommons Algorithmic Efficiency is a benchmark and competition measuring neural network training speedups due to algorithmic improvements in both training algorithms and models.
https://mlcommons.org/en/groups/research-algorithms/
Apache License 2.0
315
stars
59
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
add instructions to install Cloud Ops Agent for monitoring
#771
BharatKatyal
opened
1 week ago
1
Dev -> main
#770
priyakasimbeg
closed
1 week ago
1
cosmetic and functional fixes to scoring code
#769
priyakasimbeg
closed
1 week ago
1
Scoring funcitonal fixes and cosmetic fixes
#768
priyakasimbeg
closed
1 week ago
2
Control over batch-norm running_mean/var buffers
#767
adefazio
opened
1 month ago
0
Scoring team 11 nadamp
#766
priyakasimbeg
closed
1 month ago
1
Dev -> main
#765
priyakasimbeg
closed
1 month ago
1
fix in study index for script
#764
priyakasimbeg
closed
1 month ago
1
Update run_workloads.py
#763
priyakasimbeg
closed
1 month ago
1
Wrong return type in librispeech model_fn
#762
Niccolo-Ajroldi
opened
2 months ago
0
Dev -> main
#761
priyakasimbeg
closed
2 months ago
1
Announcement update
#760
priyakasimbeg
closed
2 months ago
1
Add support for `.eval()`
#759
netw0rkf10w
opened
2 months ago
1
Support for optimizers that require an `.eval()` step
#758
netw0rkf10w
opened
3 months ago
3
Dev -> Main
#757
priyakasimbeg
closed
3 months ago
1
Fix WMT variant targets
#756
priyakasimbeg
closed
3 months ago
1
Allow DDP checkpointing
#755
pomonam
closed
3 months ago
1
Example fails at torch.compile
#754
ClashLuke
opened
3 months ago
0
Models will always be initialized without dropout layers in self-tuning ruleset
#753
georgedahl
opened
3 months ago
0
[For review only] Self tuning submission
#752
priyakasimbeg
closed
3 months ago
1
[WIP] mismatch between old data criteo download format
#751
tfaod
closed
3 months ago
1
Clarify ported submissions in the technical documentation
#750
fsschneider
closed
3 months ago
1
dev -> main
#749
priyakasimbeg
closed
3 months ago
1
override imagenet_resnet silu and gelu bsz in configs
#748
priyakasimbeg
closed
3 months ago
1
[do not merge] variants tests
#747
priyakasimbeg
opened
3 months ago
1
Dev -> main
#746
priyakasimbeg
closed
3 months ago
1
update cl
#745
priyakasimbeg
closed
3 months ago
1
OOM with ImageNet ResNet SiLU/GELU Workloads in PyTorch
#744
hjmshi
closed
3 months ago
2
dev -> main
#743
priyakasimbeg
closed
3 months ago
1
Lint fixes
#742
priyakasimbeg
closed
3 months ago
1
Resnet DDP Warning: grad strides do not match bucket view strides
#741
Niccolo-Ajroldi
opened
3 months ago
3
Fix path in helper config for running experiments in bulk.
#740
priyakasimbeg
closed
3 months ago
1
[fix] random_utils.py to `_signed_to_unsigned`
#739
tfaod
closed
3 months ago
1
Finalize variants targets
#738
priyakasimbeg
closed
3 months ago
1
Dev -> main
#737
priyakasimbeg
closed
3 months ago
1
set num_workers for librispeech back to 4
#736
priyakasimbeg
closed
3 months ago
1
Dev -> main
#735
priyakasimbeg
closed
3 months ago
1
update change log
#734
priyakasimbeg
closed
3 months ago
1
dev -> main
#733
priyakasimbeg
closed
3 months ago
1
Incorrect Imagenet evals with pytorch_eval_num_workers > 0
#732
priyakasimbeg
opened
3 months ago
2
Dev -> main
#731
priyakasimbeg
closed
3 months ago
1
Submission instructions
#730
priyakasimbeg
closed
3 months ago
1
Dev -> main
#729
priyakasimbeg
closed
3 months ago
1
Add submission instructions
#728
priyakasimbeg
closed
3 months ago
1
update speech variants target setting points
#727
priyakasimbeg
closed
3 months ago
1
Add warning for PyTorch data loader num_workers flag.
#726
priyakasimbeg
closed
3 months ago
1
Skip eval on train and test for self-reporting results
#725
Niccolo-Ajroldi
opened
3 months ago
1
Dependency fix
#724
priyakasimbeg
closed
3 months ago
1
set default eval num workers for pytorch data loaders to 0
#723
priyakasimbeg
closed
3 months ago
1
Wandb version fix
#722
priyakasimbeg
closed
3 months ago
1
Next