Skip to content
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 datamodule for GID-15 dataset #928

Merged
merged 6 commits into from
Dec 30, 2022

Conversation

nilsleh
Copy link
Collaborator

@nilsleh nilsleh commented Dec 3, 2022

This PR adds a datamodule for the GID-15 based on #855 with the discussed random crop logic in #876.

Closes #855

@github-actions github-actions bot added datamodules PyTorch Lightning datamodules datasets Geospatial or benchmark datasets documentation Improvements or additions to documentation testing Continuous integration testing labels Dec 3, 2022
@adamjstewart adamjstewart added this to the 0.4.0 milestone Dec 9, 2022
@adamjstewart adamjstewart marked this pull request as draft December 20, 2022 17:41
@adamjstewart
Copy link
Collaborator

This will need the same treatment as #974

@github-actions github-actions bot added the scripts Training and evaluation scripts label Dec 30, 2022
adamjstewart
adamjstewart previously approved these changes Dec 30, 2022
@adamjstewart adamjstewart marked this pull request as ready for review December 30, 2022 16:58
adamjstewart
adamjstewart previously approved these changes Dec 30, 2022
@@ -88,6 +88,8 @@ filterwarnings = [
# https://github.com/lanpa/tensorboardX/issues/653
# https://github.com/lanpa/tensorboardX/pull/654
"ignore:Call to deprecated create function:DeprecationWarning:tensorboardX",
# https://github.com/kornia/kornia/issues/777
"ignore:Default upsampling behavior when mode=bilinear is changed to align_corners=False since 0.4.0:UserWarning:torch.nn.functional",
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is only needed because our tests are trying to crop a 2x2 patch from a 1x1 image, which obviously requires resizing. If someone ever implements a data.py for GID-15 and increases the image size, this can be removed.

@adamjstewart adamjstewart merged commit 2bf1a36 into microsoft:main Dec 30, 2022
yichiac pushed a commit to yichiac/torchgeo that referenced this pull request Apr 29, 2023
* add datamodule with crop logic

* remove print and fix batch_size

* typo

* Use Kornia augmentations

* Style

* Ignore warning

Co-authored-by: Adam J. Stewart <ajstewart426@gmail.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
datamodules PyTorch Lightning datamodules datasets Geospatial or benchmark datasets documentation Improvements or additions to documentation scripts Training and evaluation scripts testing Continuous integration testing
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Random sized patches support for other non-geospatial datamodules
2 participants