-
Notifications
You must be signed in to change notification settings - Fork 429
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
NF: Updating EVAC+ model and adding util function #2963
Conversation
Hello @pjsjongsung, Thank you for updating !
Comment last updated at 2023-11-15 18:07:44 UTC |
cb2c8a7
to
6c63841
Compare
Codecov Report
Additional details and impacted files@@ Coverage Diff @@
## master #2963 +/- ##
==========================================
- Coverage 81.71% 81.70% -0.01%
==========================================
Files 147 147
Lines 20484 20540 +56
Branches 3277 3285 +8
==========================================
+ Hits 16739 16783 +44
- Misses 2913 2924 +11
- Partials 832 833 +1
|
I think the PR looks ok? The failures seems to be coming from unrelated tests that uses numpy random, which I plan to fix through #2964 |
6c63841
to
ce136b5
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hi @pjsjongsung,
see below an initial reveiw before I test the code. Thank you for this!
60e2b69
to
4a2a68b
Compare
The comments should be all resolved! |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hi @pjsjongsung,
This PR seems ready to go. See below my last comments.
This commit changes the EVAC+ model and its default weights for better brain extraction output. It also adds an option to only select the largest foreground and remove its holes, so that most prediction noises can be removed. Slice wise filling holes were added as well for more desired results. Utils functions have changed to use voxsize, and tests have changed accordingly as well.
4a2a68b
to
529aa01
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you @pjsjongsung,
LGTM, merging
This commit changes the EVAC+ model and its default weights for better brain extraction output. It also adds an option to only select the largest foreground and remove its holes, so that most prediction noises can be removed.