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 aten mkldnn view operator #19209
Conversation
Differential Revision: D14894545 Differential Version: 79209391
Differential Revision: D14894545 Differential Version: 79296739
Differential Revision: D14894545 Differential Version: 79299280
Differential Revision: D14894545 Differential Version: 79689951
Differential Revision: D14894545 Differential Version: 79698399
Differential Revision: D14894545 Differential Version: 79729548
Differential Revision: D14894545 Differential Version: 80386620
Differential Revision: D14894545 Differential Version: 80533702
Differential Revision: D14894545 Differential Version: 8053430
Differential Revision: D14894545 Differential Version: 80534907
Differential Revision: D14894545 Differential Version: 80541663
Differential Revision: D14894545 Differential Version: 80564049
Differential Revision: D14894545 Differential Version: 80580143
Differential Revision: D14894545 Differential Version: 80683618
namespace at { | ||
namespace native { | ||
|
||
Tensor mkldnn_view(const Tensor& self, IntArrayRef size) { |
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.
nit: I guess you can move it out of the #ifdef as the error is regardless of whether mkldnn is compiled in
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.
The error messages are different though so I would prefer to keep two these two places.
@@ -1517,6 +1517,12 @@ | |||
variants: function, method | |||
device_guard: False | |||
|
|||
- func: mkldnn_reshape(Tensor self, int[] shape) -> Tensor | |||
device_guard: False | |||
requires_tensor: True |
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.
you don't need requires_tensor
if you have dispatch
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.
yep I know this, just for documentation purpose :-)
Differential Revision: D14894545 Differential Version: 80773734
Differential Revision: D14894545 Differential Version: 80780047
Differential Revision: D14894545 Differential Version: 80785118
Differential Revision: D14894545 Differential Version: 80799728
Summary: Pull Request resolved: pytorch/pytorch#19209 Reviewed By: dzhulgakov Differential Revision: D14894545 fbshipit-source-id: 69455184811de1d1444b5d494e4a9d8c83301431
This pull request has been merged in 48b81da. |
Summary: Pull Request resolved: pytorch#19209 Reviewed By: dzhulgakov Differential Revision: D14894545 fbshipit-source-id: 69455184811de1d1444b5d494e4a9d8c83301431
Stack:
:white_circle: #19633 Add is_mkldnn to at::Tensor 💚
:white_circle: #19204 Add aten mkldnn conv2d operator 💚
:white_circle: #19205 Add aten mkldnn ops: relu, max_pool2d and avg_pool2d 💚
:white_circle: #19206 Add aten mkldnn batch_norm operator 💚
:white_circle: #19207 Add aten mkldnn add operator 💚
:black_circle: #19209 Add aten mkldnn view operator 💚
:white_circle: #19210 Add aten mkldnn linear operator 💚
:white_circle: #19648 Adjust resnext run script 💚
Pull Request resolved: #19209
Differential Revision: D14894545