GUI selection for SGD/RAdam Training#1022
Merged
carsen-stringer merged 2 commits intoMouseLand:mainfrom Oct 22, 2024
Merged
Conversation
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
On the image.sc forum, user Brian Northan and I have found questions regarding zero masks being found during Cellpose GUI model training. Issues reported: 1, 2, and 3.
We reason that the SGD solver might be overly restrictive. Based on Brian's and my testing here, it seems that at least at low epoch numbers, certain sets of training images will result in 0 masks while using SGD.
This PR also addresses/closes issues #947 and #1017
As such, this PR adds a QCheckBox which defaults to SGD, but allows users to choose the option of the RAdam solver. It follows the same pattern as other parameters and I was able to use it successfully.