This repository has been archived by the owner on Mar 21, 2024. It is now read-only.
Fix DeepMIL metrics bug #674
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
maxilse
previously approved these changes
Feb 28, 2022
maxilse
approved these changes
Feb 28, 2022
|
For my own understanding - why the hi-ml update is needed for this PR? |
|
No other crucial changes in hi-ml related to this PR, but thought it prudent to keep it up-to-date (in particular both containing the same fixes). I can revert it if you prefer. |
| for key in thresholded_metrics_keys} | ||
|
|
||
| for key in thresholded_metrics_keys: | ||
| assert not torch.allclose(results_low_threshold[key], results_high_threshold[key]), \ |
There was a problem hiding this comment.
the default tolerance is quite low, can make sense to pass it a larger number given here we're interested in checking these results ARE different?
|
No need to revert - I was just trying to understand. |
vale-salvatelli
approved these changes
Feb 28, 2022
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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.
Fixes a bug in DeepMILModule whereby predicted hard labels were being passed to classification metrics, instead of the intended continuous probability scores. This PR also adds tests to ensure this won't happen anymore.