Skip to content

Fix ShardDataloader and embedding_grad_kernel bugs #66096

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

Conversation

From00
Copy link
Contributor

@From00 From00 commented Jul 16, 2024

PR Category

Auto Parallel

PR Types

Bug fixes

Description

Pcard-76459

  1. ShardDataloader的_get_batch接口在is_dataset_splitted=True且数据是dict类型的情况下,支持input_key=None的情况
  2. 修复embedding_grad_kernel在词表较大时爆int的问题

Copy link

paddle-bot bot commented Jul 16, 2024

你的PR提交成功,感谢你对开源项目的贡献!
请关注后续CI自动化测试结果,详情请参考Paddle-CI手册
Your PR has been submitted. Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

@From00 From00 changed the title Add split inferspmd rule for dygraph and fix some bugs Fix ShardDataloader and embeding_grad_kernel bugs Jul 16, 2024
@From00 From00 changed the title Fix ShardDataloader and embeding_grad_kernel bugs Fix ShardDataloader and embedding_grad_kernel bugs Jul 16, 2024
Copy link
Contributor

@deepllz deepllz left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

Copy link
Contributor

@zhiqiu zhiqiu left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@From00 From00 merged commit 3399250 into PaddlePaddle:develop Jul 18, 2024
30 of 31 checks passed
lixcli pushed a commit to lixcli/Paddle that referenced this pull request Jul 22, 2024
* Add split inferspmd rule for dygraph and fix shard dataloader bug

* Update

* Update
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.

3 participants