Skip to content

Conversation

@vinx13
Copy link
Member

@vinx13 vinx13 commented Jul 30, 2019

No description provided.

@vinx13 vinx13 force-pushed the feature/group_conv_cuda branch from 5114d3a to ec03729 Compare July 30, 2019 13:58
@vinx13
Copy link
Member Author

vinx13 commented Jul 30, 2019

@merrymercy @masahi @kevinthesun please help review

s[output].pragma(kernel_scope, 'unroll_explicit', cfg['unroll_explicit'].val)

N, CO, OH, OW = get_const_tuple(output.shape)
_, KH, KW, CI = get_const_tuple(kernel.shape)
Copy link
Member

Choose a reason for hiding this comment

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

Is this layout correct? Shouldn't it be CO, group, KH, KW?

Copy link
Member Author

Choose a reason for hiding this comment

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

nice catch!

@masahi masahi merged commit 11da1ca into apache:master Jul 31, 2019
@masahi
Copy link
Member

masahi commented Jul 31, 2019

thanks @vinx13, this is merged.

wweic pushed a commit to wweic/tvm that referenced this pull request Aug 9, 2019
* [TOPI][CUDA] schedule for group_conv2d

* Fix #flops
wweic pushed a commit to neo-ai/tvm that referenced this pull request Sep 6, 2019
* [TOPI][CUDA] schedule for group_conv2d

* Fix #flops
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