Closed JackDunnNZ closed 3 years ago
Merging #54 (9f367aa) into master (f644c26) will decrease coverage by
0.26%
. The diff coverage is90.00%
.
@@ Coverage Diff @@
## master #54 +/- ##
==========================================
- Coverage 93.53% 93.27% -0.27%
==========================================
Files 3 3
Lines 464 446 -18
==========================================
- Hits 434 416 -18
Misses 30 30
Impacted Files | Coverage Δ | |
---|---|---|
src/CoxNet.jl | 88.50% <ø> (-1.68%) |
:arrow_down: |
src/GLMNet.jl | 94.98% <87.50%> (-0.26%) |
:arrow_down: |
src/Multinomial.jl | 93.00% <100.00%> (ø) |
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 f644c26...9f367aa. Read the comment docs.
After #50, I saw some test failures on CI at random due to the RNG tests returning fits of different lengths in rare cases. I realized the Cox version of
glmnetcv
didn't have the same guards against the fits being different lengths.It seemed a clean fix to merge these functions together rather than having all the cross-validation logic duplicated between them.