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

Predict zero for nodata pixels on semantic segmentation #701

merged 1 commit into from Mar 4, 2019


Copy link

@lewfish lewfish commented Mar 4, 2019


This PR makes it so that a zero (don't care value) is predicted for each pixel in the RasterSource that has a NODATA value. This makes it so the NODATA outlines around a scene don't have false positive "hallucinations".

Testing Instructions

I didn't unit test this because there aren't any existing unit tests for the Tasks, and the integration tests don't test AOIs or NODATA handling, and that will take too much time right now. I added an issue for this #702 However, I did find that it has the desired effect on a client project.

Closes #703

@lewfish lewfish added the review label Mar 4, 2019
@lewfish lewfish force-pushed the lf/nodata branch from 5d0d557 to 95ac90a Mar 4, 2019
@lewfish lewfish changed the title WIP: predict 0 for nodata Predict zero for nodata pixels on semantic segmentation Mar 4, 2019
@lewfish lewfish force-pushed the lf/nodata branch from 95ac90a to df898b7 Mar 4, 2019
Copy link

@codecov codecov bot commented Mar 4, 2019

Codecov Report

No coverage uploaded for pull request base (develop@e7aa5ed). Click here to learn what that means.
The diff coverage is 0%.

Impacted file tree graph

@@            Coverage Diff             @@
##             develop     #701   +/-   ##
  Coverage           ?   71.66%           
  Files              ?      171           
  Lines              ?     8283           
  Branches           ?        0           
  Hits               ?     5936           
  Misses             ?     2347           
  Partials           ?        0
Impacted Files Coverage Δ
rastervision/task/ 22.53% <0%> (ø)

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update e7aa5ed...df898b7. Read the comment docs.

@lewfish lewfish merged commit 9f38cc9 into develop Mar 4, 2019
1 of 2 checks passed
1 of 2 checks passed
continuous-integration/travis-ci/push The Travis CI build failed
continuous-integration/travis-ci/pr The Travis CI build passed
@lewfish lewfish deleted the lf/nodata branch Mar 4, 2019
@lewfish lewfish removed the review label Mar 4, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
None yet
None yet
Linked issues

Successfully merging this pull request may close these issues.

None yet

1 participant