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

Handle "ignore" class for semantic segmentation #783

Merged
merged 8 commits into from May 23, 2019

Conversation

@lewfish
Copy link
Contributor

commented May 15, 2019

Overview

This PR makes it so we handle an "ignore" class in a more sensible way for semantic segmentation. This establishes a convention that class 0 represents an "ignore" class of pixels that should be ignored for the sake of training and evaluation. Imagery pixels with label class 0 are now set to 0 (NODATA) to prevent the model from associating good imagery with the ignore class. Chips that only contain class 0 aren't saved since they are a waste of time to train on. And, the eval does not generate metrics for class 0.

Testing

This is tested with a unit test and also in an experiment that used class 0, where debug chips were inspected for correctness.

@lewfish lewfish added the review label May 15, 2019
@lewfish lewfish force-pushed the lf/ignore-class branch from d23cc18 to 878623a May 23, 2019
@lewfish lewfish changed the title WIP: Ignore class 0 during eval Handle "ignore" class for semantic segmentation May 23, 2019
@lewfish lewfish merged commit 9d6d808 into master May 23, 2019
1 check passed
1 check passed
continuous-integration/travis-ci/pr The Travis CI build passed
Details
@lewfish lewfish deleted the lf/ignore-class branch May 23, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
1 participant
You can’t perform that action at this time.