Skip to content
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.

Sign up for GitHub

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

Jump to bottom

Cleanup of cross entropy calculation #38

Merged
merged 2 commits into from
Sep 9, 2021
Merged

Cleanup of cross entropy calculation #38

merged 2 commits into from
Sep 9, 2021

Conversation

cDenius
Copy link
Contributor

@cDenius cDenius commented Sep 1, 2021 •
edited by eileen-kuehn
Loading

Cross entropy calculation now weights results properly by number of samples.
Further, the PR includes documentation, unit tests and normalisation of predictions.

Closes #47.

@cDenius cDenius linked an issue Sep 1, 2021 that may be closed by this pull request
Cross entropy loss is divided by the number of classes and not the batch size #34
Closed
@codecov
Copy link

codecov bot commented Sep 1, 2021 •
edited
Loading

Codecov Report

Merging #38 (21eb8e9) into main (ccaff9b) will increase coverage by 5.73%.
The diff coverage is 100.00%.

Impacted file tree graph

@@            Coverage Diff             @@
##             main      #38      +/-   ##
==========================================
+ Coverage   74.54%   80.28%   +5.73%     
==========================================
  Files          15       16       +1     
  Lines        1096     1486     +390     
  Branches      105      162      +57     
==========================================
+ Hits          817     1193     +376     
- Misses        275      288      +13     
- Partials        4        5       +1
Impacted Files Coverage Δ
maskit/utils.py 100.00% <100.00%> (+100.00%) ⬆️
tests/test_utils.py 100.00% <100.00%> (ø)
main.py 0.00% <0.00%> (ø)
maskit/masks.py 100.00% <0.00%> (ø)
tests/test_masks.py 100.00% <0.00%> (ø)
maskit/circuits.py 76.19% <0.00%> (+76.19%) ⬆️

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 0313002...21eb8e9. Read the comment docs.

This was referenced Sep 9, 2021
Closed
Merged
@eileen-kuehn
Documentation, test and normalization for cross entropy calculation (#45
21eb8e9 
)

* add documentation for cross entropy

* test cross entropy calculation with sklearn log_loss

* add normalization for predictions
@eileen-kuehn eileen-kuehn changed the title cross entropy loss is now divided by batch size Cleanup of cross entropy calculation Sep 9, 2021
eileen-kuehn
eileen-kuehn approved these changes Sep 9, 2021
View reviewed changes
Copy link
Member

@eileen-kuehn eileen-kuehn left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Great stuff! 🥇

@eileen-kuehn eileen-kuehn merged commit e135be3 into main Sep 9, 2021
@eileen-kuehn eileen-kuehn deleted the ce_fix branch September 9, 2021 10:28
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@eileen-kuehn eileen-kuehn eileen-kuehn approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
2 participants
@cDenius @eileen-kuehn