search

mlr-org / mlr3learners

Recommended learners for mlr3
https://mlr3learners.mlr-org.com
GNU Lesser General Public License v3.0
91 stars 16 forks source link

Add surv glmnet #130

Closed RaphaelS1 closed 4 years ago

RaphaelS1 commented 4 years ago

Param check is passing but others seem to be failing because of classif xgboost and not these learners

RaphaelS1 commented 4 years ago

Thanks. Master is green so I'm wondering why this PR fails due to xgboost if there were no changes?

The only thing I can think of is the mlr3proba helpers, but these are proba specific so I don't see how they could affect xgboost. Also the error doesn't seem related to that. I had the latest version of mlr3learners and installed latest mlr3 so I don't really understand either. I can try and dig around more though to find out

RaphaelS1 commented 4 years ago

The problem is that it's currently calling the old version of mlr3proba before autotests were merged. I won't add the remotes, I'll publish proba to CRAN tomorrow then fix this

pat-s commented 4 years ago

sounds good!

RaphaelS1 commented 4 years ago

Checks should all pass now but note that this now requires mlr3 0.3.0

codecov-commenter commented 4 years ago

Codecov Report

Merging #130 into master will decrease coverage by 0.11%. The diff coverage is 95.77%.

Impacted file tree graph

@@            Coverage Diff             @@
##           master     #130      +/-   ##
==========================================
- Coverage   96.88%   96.76%   -0.12%     
==========================================
  Files          21       23       +2     
  Lines        1219     1361     +142     
==========================================
+ Hits         1181     1317     +136     
- Misses         38       44       +6
Impacted Files Coverage Δ
R/LearnerClassifCVGlmnet.R 95.65% <ø> (ø)
R/LearnerSurvCVGlmnet.R 94.73% <94.73%> (ø)
R/LearnerSurvGlmnet.R 96.87% <96.87%> (ø)
R/zzz.R 78.57% <100.00%> (+1.64%) :arrow_up:

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 fd6f10f...773f2bf. Read the comment docs.

RaphaelS1 commented 4 years ago

All checks now passing @pat-s