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 GeoNRW dataset #2209

Merged
merged 20 commits into from
Aug 27, 2024
Merged

Add GeoNRW dataset #2209

merged 20 commits into from
Aug 27, 2024

Conversation

nilsleh
Copy link
Collaborator

@nilsleh nilsleh commented Aug 6, 2024

This PR adds the GeoNRW dataset:

As it is licensed under CC-BY-4.0, I put it on HF for easier usage.

TODOS:

  • write unit tests
  • automatic download and verify
  • check plotting because it seems off, although I followed this
    Screenshot from 2024-08-06 14-28-02

@github-actions github-actions bot added documentation Improvements or additions to documentation datasets Geospatial or benchmark datasets datamodules PyTorch Lightning datamodules labels Aug 6, 2024
@adamjstewart adamjstewart added this to the 0.6.0 milestone Aug 6, 2024
@github-actions github-actions bot added the testing Continuous integration testing label Aug 7, 2024
@nilsleh nilsleh marked this pull request as draft August 8, 2024 12:40
@nilsleh
Copy link
Collaborator Author

nilsleh commented Aug 20, 2024

@isaaccorley Could you take a look at the plotting, and possibly find a mistake based on their code and this implementation, since there seem to be some visual discrepancies between optical and segmentation image?

@nilsleh nilsleh marked this pull request as ready for review August 20, 2024 07:42
Copy link
Collaborator

@adamjstewart adamjstewart left a comment

Choose a reason for hiding this comment

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

Will investigate the plotting code and get back to you.

torchgeo/datasets/geonrw.py Show resolved Hide resolved
docs/api/non_geo_datasets.csv Outdated Show resolved Hide resolved
tests/data/geonrw/data.py Show resolved Hide resolved
tests/data/geonrw/data.py Show resolved Hide resolved
tests/data/geonrw/data.py Show resolved Hide resolved
torchgeo/datasets/geonrw.py Outdated Show resolved Hide resolved
torchgeo/datasets/geonrw.py Outdated Show resolved Hide resolved
torchgeo/datasets/geonrw.py Show resolved Hide resolved
torchgeo/datasets/geonrw.py Outdated Show resolved Hide resolved
torchgeo/datasets/geonrw.py Show resolved Hide resolved
@adamjstewart
Copy link
Collaborator

adamjstewart commented Aug 27, 2024

As it is licensed under CC-BY-4.0, I put it on HF for easier usage.

You need to provide attribution (a link to the original version/authors) and list any modifications you made to the dataset (preprocessing?). Can you update the HF README?

See https://creativecommons.org/licenses/by/4.0/deed.en

@adamjstewart
Copy link
Collaborator

Note to self: this dataset requires PIL built with jpeg2000 support, not just regular jpeg. We can update the Spack recipe to include this.

@adamjstewart
Copy link
Collaborator

I successfully downloaded, loaded, and plotted the data. Seems like all TODOs are now complete.
geonrw

@adamjstewart adamjstewart merged commit 2d6e27e into microsoft:main Aug 27, 2024
19 checks passed
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 testing Continuous integration testing
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants