Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add Resnet50 for bf16 mixed precision #5316

Open
wants to merge 1 commit into
base: dev-static
Choose a base branch
from

Conversation

AshburnLee
Copy link

@AshburnLee AshburnLee commented Jun 9, 2021

PR功能

PR支持了GPU单卡 Resnet50 的bfloat16混合精度训练。

方法

  1. 添加参数use_amp_bf16 & use_pure_bf16 并设false为初始值。
  2. train.py中调用接口decorate_bf16(),并指定使用bfloat16精度的op,将其放入custom_bf16_list中。
  3. 添加执行bf16 混合精度训练的脚本Resnet50_bf16.sh

@CLAassistant
Copy link

CLA assistant check
Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
You have signed the CLA already but the status is still pending? Let us recheck it.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants