Croppad inverse transform#1737
Conversation
Signed-off-by: Richard Brown <[email protected]>
Signed-off-by: Richard Brown <[email protected]>
Signed-off-by: Richard Brown <[email protected]>
Signed-off-by: Richard Brown <[email protected]>
This reverts commit 40322d7. Signed-off-by: Richard Brown <[email protected]>
Nic-Ma
left a comment
There was a problem hiding this comment.
Looks good to me.
Thanks.
Signed-off-by: Richard Brown <[email protected]>
Signed-off-by: Richard Brown <[email protected]>
… croppad_inverse_transform Signed-off-by: Richard Brown <[email protected]>
Signed-off-by: Richard Brown <[email protected]>
|
/integration-test |
…sform Signed-off-by: Richard Brown <[email protected]>
Signed-off-by: Richard Brown <[email protected]>
…sform Signed-off-by: Richard Brown <[email protected]>
Signed-off-by: Richard Brown <[email protected]>
Signed-off-by: Richard Brown <[email protected]>
|
@wyli if you're happy with the PR now, would you mind enabling auto-merge? |
wyli
left a comment
There was a problem hiding this comment.
thanks, this looks good to me!
|
there's an assertion error that occurs randomly...
Failed: RandSpatialCropd1D odd Compose. Mean diff = 0.47619047619047616 (expected <= 0), unmodified diff: 2.0952380952380953 |
|
This could happen if the part that is randomly cropped includes non-zero voxels. But given that the seed is used during |
|
yes, strange that I couldn't replicate the issue locally on Linux or mac os. I've seen the errors from the CI a few times, the mean diff is always 0.4761... though |
Croppad inverse transforms. All lossless (thus all the 0's in
test_inverse.py).#1515
Status
Ready
Types of changes
./runtests.sh --codeformat --coverage../runtests.sh --quick.make htmlcommand in thedocs/folder.