Elias Wahl 27613dd881 MLPerf BERT: Main training loop (#4288) ... * BERT language modeling head + trunc normal initializers * add train loop + helpers * shuffle in dataloaders + slight changes in main loop * beam change * Minor changes * random.shuffle * HParam update * Use deque for dataloader * wandb bert project name * half fixes * BENCHMARK + remove epoch * cast + print() --------- Co-authored-by: chenyu <chenyu@fastmail.com>		2024-04-29 14:35:27 -04:00
..
.gitignore	mock mnist data for imagenet trainer (#4095)	2024-04-06 08:08:40 -07:00
__init__.py	remove numpy from ops_torch (#3124)	2024-01-14 22:46:57 -05:00
coco.py	move to new cached fetch (#2493)	2023-11-28 17:36:55 -08:00
fake_imagenet_from_mnist.py	mock mnist data for imagenet trainer (#4095)	2024-04-06 08:08:40 -07:00
imagenet.py	mock mnist data for imagenet trainer (#4095)	2024-04-06 08:08:40 -07:00
imagenet_download.py	move to new cached fetch (#2493)	2023-11-28 17:36:55 -08:00
kits19.py	[MLPerf] UNet3D dataloader (#4343)	2024-04-28 22:34:18 -04:00
librispeech.py	Update file paths (#1179)	2023-07-07 18:41:58 -07:00
openimages.py	Fix missing import. (#3666)	2024-03-09 14:55:23 -08:00
preprocess_imagenet.py	move dtypes to dtype.py (#2964)	2024-01-01 14:58:48 -08:00
sops.gz	update kernel dataset to exclude the disktensor ones (#3651)	2024-03-07 17:35:19 -05:00
squad.py	move to new cached fetch (#2493)	2023-11-28 17:36:55 -08:00
wikipedia.py	MLPerf BERT: Main training loop (#4288)	2024-04-29 14:35:27 -04:00
wikipedia_download.py	Wikipedia download script for MLPerf BERT training (#4202)	2024-04-17 16:34:57 -04:00