Skip to content

Conversation

@swinersha
Copy link
Collaborator

This is an experimental branch where:

  • K is defined as all the pixels within the project polygon (instead of a subsample).
  • Find pairs has been modified so that a random 1% sample of K is matched to a random 10% subset of M.
  • Find pairs also breaks M into smaller chunks using 9 clusters produced using k-means clustering.

At the moment we are looking for feedback and plan to do some experiments to compare the resulting matches and additionality with those from the current version on main. There are definitely places where the code can be improved.

# RasterLayer.layer_from_file(os.path.join(ecoregions_directory_path, filename)) for filename in
# glob.glob("*.tif", root_dir=ecoregions_directory_path)
# ], name="ecoregions")
ecoregions = RasterLayer.layer_from_file(ecoregions_directory_path)
Copy link
Contributor

Choose a reason for hiding this comment

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

This looks like somehow @swinersha has based his work off the work-in-progress branch I have for per-project optimisation, rather than from main. But he doesn't seem have picked up any other changes oddly enough. I was worried I'd accidentally committed something I shouldn't to main, but that doesn't appear to be the case.

swinersha and others added 4 commits August 7, 2024 13:34
Removed the problematic lines for ecoregion and access
Uncommented lines needed for ecoregion and access
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants