Skip to content

fixed ARI (fixes #225 & #226)#227

Merged
alyst merged 4 commits intoJuliaStats:masterfrom
wildart:cm
Mar 19, 2023
Merged

fixed ARI (fixes #225 & #226)#227
alyst merged 4 commits intoJuliaStats:masterfrom
wildart:cm

Conversation

@wildart
Copy link
Copy Markdown
Contributor

@wildart wildart commented Dec 25, 2021

  • added pair confusion matrix (CM) calculations
  • fixed ARI calculation using CM

@codecov-commenter
Copy link
Copy Markdown

codecov-commenter commented Dec 25, 2021 •
edited
Loading

Codecov Report

Base: 95.18% // Head: 95.12% // Decreases project coverage by -0.05% ⚠️

Coverage data is based on head (fc57919) compared to base (82821e8).
Patch coverage: 90.47% of modified lines in pull request are covered.

❗ Current head fc57919 differs from pull request most recent head 945c537. Consider uploading reports for the commit 945c537 to get more accurate results

Additional details and impacted files 
@@            Coverage Diff             @@
##           master     #227      +/-   ##
==========================================
- Coverage   95.18%   95.12%   -0.06%     
==========================================
  Files          16       17       +1     
  Lines        1328     1333       +5     
==========================================
+ Hits         1264     1268       +4     
- Misses         64       65       +1
Impacted Files Coverage Δ
src/confusion.jl 83.33% <83.33%> (ø)
src/randindex.jl 100.00% <100.00%> (+5.55%) ⬆️

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

@wildart
Copy link
Copy Markdown
Contributor Author

wildart commented Dec 25, 2021

As rand behavior changed in recent julia versions, the rework of RNG handling or a test review is required.

@wildart wildart mentioned this pull request Dec 25, 2021
Open
@pcjentsch
Copy link
Copy Markdown

pcjentsch commented Sep 8, 2022

This PR passes the relevant tests, can it be merged?

This is an old bug that has had other PRs to fix it: #99

@ararslan ararslan closed this Sep 10, 2022
@ararslan ararslan reopened this Sep 10, 2022
@ararslan
Copy link
Copy Markdown
Member

ararslan commented Sep 10, 2022

Looks like tests are passing except for 32-bit Linux. I wonder if the forced conversion to Int in confusion means that something in randindex overflows when Int is only half as big.

pcjentsch
pcjentsch reviewed Sep 11, 2022
View reviewed changes
Comment thread test/randindex.jl Outdated
@wildart
fixed ARI (fixes JuliaStats#225 & JuliaStats#226)
a661450 
- added pair confusion matrix (CM) calculations
- fixed ARI calculation using CM
@wildart
fix test for x86
f773f92
@pcjentsch
Copy link
Copy Markdown

pcjentsch commented Sep 23, 2022 •
edited
Loading

Hm, not sure why that change broke the test. Sorry about that, I did test it locally before suggesting it and it was fine.

alyst
alyst reviewed Sep 23, 2022
View reviewed changes
Copy link
Copy Markdown
Member

@alyst alyst left a comment

Choose a reason for hiding this comment

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

Thanks for the PR!
Looks good. Could you please implement the suggestions (which are basically just formatting improvements)?

Comment thread doc/source/validate.md Outdated
Comment thread doc/source/validate.md Outdated
Comment thread src/confusion.jl Outdated
Comment thread src/confusion.jl Outdated
Comment thread src/confusion.jl Outdated
Comment thread src/confusion.jl
Comment thread src/confusion.jl Outdated
Comment thread src/randindex.jl Outdated
Comment thread src/randindex.jl Outdated
Comment thread test/randindex.jl Outdated
@alyst alyst merged commit 5d1ad38 into JuliaStats:master Mar 19, 2023
@alyst
Copy link
Copy Markdown
Member

alyst commented Mar 19, 2023

@wildart So sorry I overlooked your updates! Thank you!

alyst added a commit that referenced this pull request Mar 20, 2023
@alyst
randindex(): fix int overflow
56bafb0 
for very large clusterings the agreement/disagreement counts are
very large, so we have to switch to float when multiplying them

fixes #225
enhances #227
@alyst alyst mentioned this pull request Mar 20, 2023
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@alyst alyst alyst left review comments

+1 more reviewer

@pcjentsch pcjentsch pcjentsch left review comments

Reviewers whose approvals may not affect merge requirements

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

5 participants

@wildart @codecov-commenter @pcjentsch @ararslan @alyst