iperov 35877dbfd7 AMP, SAEHD: added option use_fp16: Increases training/inference speed, reduces model size. Model may crash. Enable it after 1-5k iters. ...

AMP: removed lr_dropout, now it is enabled by default;

2021-07-14 23:03:54 +04:00

..

archis

AMP, SAEHD: added option use_fp16: Increases training/inference speed, reduces model size. Model may crash. Enable it after 1-5k iters.

2021-07-14 23:03:54 +04:00

initializers

global refactoring and fixes,

2020-03-13 08:09:00 +04:00

layers

fix

2021-06-10 14:42:58 +04:00

models

AMP, SAEHD: added option use_fp16: Increases training/inference speed, reduces model size. Model may crash. Enable it after 1-5k iters.

2021-07-14 23:03:54 +04:00

ops

AMP fix

2021-05-30 09:24:23 +04:00

optimizers

leras: added AdaBelief optimizer

2020-12-11 15:47:47 +04:00

__init__.py

DFL-2.0 initial branch commit

2020-01-21 18:43:39 +04:00

device.py

fix for colab

2021-04-22 21:12:42 +04:00

nn.py

added AMD/Intel cards support via DirectX12 ( DirectML backend )

2021-04-22 18:19:15 +04:00