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 ReforesTree dataset #582

Merged
merged 10 commits into from
Jul 9, 2022
Merged

Add ReforesTree dataset #582

merged 10 commits into from
Jul 9, 2022

Conversation

nilsleh
Copy link
Collaborator

@nilsleh nilsleh commented Jun 15, 2022

This PR adds ReforestTree dataset for tree crown detection and aboveground biomass (AGB) estimation from the ReforesTree paper. The __getitem__ method returns images, boxes, tree species labels, and AGB values.

Features:

  • high resolution rgb imagery at 2cm/px and 4000x4000px
  • generated and manually adjusted and matched bounding boxes as well as field measurements of AGB

Issue:

  • the data is hosted on DropBox and the automatic download pipeline is not working with the download link (Access forbidden or not a zip file)

Example:

@github-actions github-actions bot added datasets Geospatial or benchmark datasets documentation Improvements or additions to documentation testing Continuous integration testing labels Jun 15, 2022
@adamjstewart adamjstewart added this to the 0.3.0 milestone Jun 16, 2022
calebrob6
calebrob6 previously approved these changes Jun 18, 2022
Copy link
Member

@calebrob6 calebrob6 left a comment

Choose a reason for hiding this comment

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

Looks great to me! I like the object detection plotting.

@nilsleh
Copy link
Collaborator Author

nilsleh commented Jun 18, 2022

Do you have an opinion about handling the download with Dropbox? Currently, it is not working and just throwing errors, so maybe there is a way around it or I write a description in the class for download instructions?

@calebrob6
Copy link
Member

calebrob6 commented Jun 29, 2022

It looks like the link you are using is auto generated by dropbox when you visit the current dataset download page -- https://www.dropbox.com/scl/fo/qzlq48k06dbksvrl3kk1o/h?dl=0&rlkey=hpq3r63b1tbpqa02f5tyh8l6s. I pinged to authors to see if they will host a single zip file for download (which we can easily download) / we can rehost on Zenodo.

docs/api/datasets.rst Show resolved Hide resolved
tests/data/reforestree/data.py Show resolved Hide resolved
torchgeo/datasets/reforestree.py Outdated Show resolved Hide resolved
torchgeo/datasets/reforestree.py Outdated Show resolved Hide resolved
torchgeo/datasets/reforestree.py Outdated Show resolved Hide resolved
torchgeo/datasets/reforestree.py Outdated Show resolved Hide resolved
torchgeo/datasets/reforestree.py Outdated Show resolved Hide resolved
torchgeo/datasets/reforestree.py Outdated Show resolved Hide resolved
@adamjstewart
Copy link
Collaborator

Do you have an opinion about handling the download with Dropbox? Currently, it is not working and just throwing errors, so maybe there is a way around it or I write a description in the class for download instructions?

If it's not working I would remove all the download logic for now. We can add it in a new release if the authors are willing to host elsewhere.

@calebrob6
Copy link
Member

calebrob6 commented Jul 4, 2022 via email

@adamjstewart
Copy link
Collaborator

Let's wait then. I don't think we're going to finish 0.3.0 this weekend, going to have to push it to next weekend until we have a chance to resolve these CI issues.

@adamjstewart
Copy link
Collaborator

adamjstewart commented Jul 9, 2022

Is the MD5 the same?

EDIT: downloading now to check

@calebrob6
Copy link
Member

calebrob6 commented Jul 9, 2022 via email

@calebrob6
Copy link
Member

calebrob6 commented Jul 9, 2022

Should be good to go now

Confirmed that it works:
image

@adamjstewart
Copy link
Collaborator

I have a bunch of edits, but I'll just push them myself

adamjstewart
adamjstewart previously approved these changes Jul 9, 2022
torchgeo/datasets/reforestree.py Outdated Show resolved Hide resolved
adamjstewart
adamjstewart previously approved these changes Jul 9, 2022
@adamjstewart adamjstewart merged commit 7cd28ca into microsoft:main Jul 9, 2022
@adamjstewart adamjstewart mentioned this pull request Jul 11, 2022
yichiac pushed a commit to yichiac/torchgeo that referenced this pull request Apr 29, 2023
* add ReforesTree dataset

* fix failing test

* suggested changes

* Update download URL

* Change zipfile name

* Minor fixes

* Remove f-string

* Fix dtype, remove unnecessary conversion

Co-authored-by: Caleb Robinson <calebrob6@gmail.com>
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
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.

None yet

3 participants