Closed btrotta-bom closed 5 months ago
Attention: 1 lines
in your changes are missing coverage. Please review.
Comparison is base (
14723f3
) 98.40% compared to head (42c1bb6
) 98.40%.
Files | Patch % | Lines |
---|---|---|
improver/calibration/rainforest_calibration.py | 98.59% | 1 Missing :warning: |
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
In order to maintain a backlog of relevant PRs, we automatically label them as stale after 60 days of inactivity.
If this PR is still important to you, then please comment on this PR and the stale label will be removed.
Otherwise this PR will be automatically closed in 30 days time.
This is still current.
Scheduled tests are failing with the latest
env. We should pin the treelite package in latest.yml
to resolve.
Scheduled tests are failing with the
latest
env. We should pin the treelite package inlatest.yml
to resolve.
Done
Thanks for reviewing @benowen-bom and @bomRob, all changes are merged now.
I realised after merging that the fix to envs/latest.yml
to pin the treelite package wasn't included here. I'll add a PR tomorrow to enact this change.
Add an option to speed up rainforests prediction.
The improved efficiency is achieved by observing that for sets of grid points having very similar values for their features, the features will be binned by LightGBM in the same way for each point, and therefore the predicted value of each point will be the same. Thus we only need to predict once for the whole set of points. This yields a reduction of around 20-40% in processing time for gridded forecasts (depending on file).
Description
Testing: