Skip to content

[XPU] support xpu mask adaptive output without embedding; test=develop#9970

Merged
zhupengyang merged 1 commit intoPaddlePaddle:developfrom
newway:vsl_v3
Feb 28, 2023
Merged

[XPU] support xpu mask adaptive output without embedding; test=develop#9970
zhupengyang merged 1 commit intoPaddlePaddle:developfrom
newway:vsl_v3

Conversation

@newway
Copy link
Collaborator

@newway newway commented Feb 6, 2023

PR devices

PR types

PR changes

Description

@paddle-bot
Copy link

paddle-bot bot commented Feb 6, 2023

Thanks for your contribution!

Copy link
Collaborator

@zhupengyang zhupengyang left a comment

Choose a reason for hiding this comment

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

LGTM

Comment on lines +27 to +31
if (input_shape.size() == 2) {
// need to insert an sequence_unpad op to enable some vsl pass
// the output shape of sequence_unpad is 2dims
param_.input->Resize({1ULL, input_shape[0], input_shape[1]});
}
Copy link
Collaborator

Choose a reason for hiding this comment

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

修改input shape的做法是很trick的。如果有需求,应该在pass中增加reshape op

@zhupengyang zhupengyang merged commit 9751487 into PaddlePaddle:develop Feb 28, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants