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
Enhance BatchToSpaceND and SpaceToBatchND handlers #1386
Enhance BatchToSpaceND and SpaceToBatchND handlers #1386
Conversation
- add explicit NHWC_TO_NCHW and NCHW_TO_NHWC transposes so they can be eliminated with its counterpart in TransposeOptimizer - handle nonconst pads and crops Signed-off-by: Mateusz Tabaka <mateusz.tabaka@intel.com>
Is this ready for review + merge? |
Yes |
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.
Thanks for the contribution. I'm not entirely sure that your code is being tested. Can you double-check?
@@ -3323,7 +3323,7 @@ def func(x, z): | |||
|
|||
@check_opset_min_version(11, "SpaceToBatchND") | |||
def test_space_to_batchnd_non_const_7d(self): | |||
x_type, y_type, z_type = np.int64, np.int64, np.int64 | |||
x_type, y_type, z_type = np.float32, np.int64, np.int64 |
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 black_shapes include [2, 2]
in this case, so it appears the code is properly tested.
Signed-off-by: Mateusz Tabaka <mateusz.tabaka@intel.com>
f190a6a
to
cf9df39
Compare
eliminated with its counterpart in TransposeOptimizer
Signed-off-by: Mateusz Tabaka mateusz.tabaka@intel.com