-
Notifications
You must be signed in to change notification settings - Fork 5.7k
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 dilation for conv_trans_op #6279
Add dilation for conv_trans_op #6279
Conversation
It seems no code is using attribute dilations? |
The code is not fully implemented yet. @jacquesqiao |
1ea3e68
to
d93bbf1
Compare
e450dfb
to
479ee27
Compare
self.op_type = "conv2d_transpose_cudnn" | ||
|
||
|
||
# #cudnn v5 does not support dilation conv. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
直接注释不太合适吧?对不同CUDNN_VERSION做测试,你可以参考这里:
https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/gserver/tests/test_LayerGrad.cpp#L198
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
我觉得可以在下一个PR里面添加一个从python端获取CUDNN_VERSION的接口
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
fix #6189
fix #6423