Allow all constituent object detection losses to be logged #1716
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Overview
This PR removes
Learner.build_metric_names()
. This means that all metrics returned byLearner.train_step()
andLearner.validation_step()
can be logged without needing to be pre-specified inLearner.build_metric_names()
, which means losses from different object detection models (which can have different names) can be logged without needing to override the defaultObjectDetectionLearner
.After the changes, training logs with a FasterRCNN model look like so:
With an external model (SSD):
Other changes:
Checklist
needs-backport
label if PR is bug fix that applies to previous minor releaseNotes
N/A
Testing Instructions