-
Notifications
You must be signed in to change notification settings - Fork 45
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
Composable Data Augmentations #189
Composable Data Augmentations #189
Conversation
for more information, see https://pre-commit.ci
for more information, see https://pre-commit.ci
this seems like a breaking change. if the default boolean option is dropped, it should be scheduled for deprecation in a few cycles and this code should warn people that the boolean option will be removed in a future version. i suspect that this is not being caught by current tests, so we should add a test as well. |
Yes @satra, I will add the test and tutorial ipynb in this PR. |
for more information, see https://pre-commit.ci
for more information, see https://pre-commit.ci
Fixes #167 |
for more information, see https://pre-commit.ci
for more information, see https://pre-commit.ci
@Aakanksha-Rana - please check these changes, and do make sure that the transforms are capable of understanding what to do with the y-value (if it's a scalar, if it's image with integer labels, etc.,.)
for more information, see https://pre-commit.ci
for more information, see https://pre-commit.ci
We need to change Transforms.py functions in same format as intensity/spatial transforms to be able to take scalar or non-scalar labels. |
can this be done for this PR ? |
I will change the apply_random_transform in this PR @satra . |
for more information, see https://pre-commit.ci
for more information, see https://pre-commit.ci
for more information, see https://pre-commit.ci
Types of changes
Summary
The goal of this PR is to enable sequence of on-the-fly augmentation by giving multiple transforms as list for argument 'augment' . This is open to suggestions and modifications to make it better and robust for different transform functionalities available in spatial_transform, intensity_transform and volume.
Checklist
Acknowledgment