|
Ran
|
Jobs
2
|
Files
6
|
Run time
18min
|
Badge
README BADGES
|
push
travis-ci
<a href="https://github.com/JuliaStats/StatsModels.jl/commit/<a class=hub.com/JuliaStats/StatsModels.jl/commit/f1e2675632f2140aa891eaf5fc03ec8cc9bc92af">f1e267563<a href="https://github.com/JuliaStats/StatsModels.jl/commit/f1e2675632f2140aa891eaf5fc03ec8cc9bc92af">">Move back to DataFrames and port from Nullable to Missings (#30) This reverts commit </a><a class="double-link" href="https://github.com/JuliaStats/StatsModels.jl/commit/<a class="double-link" href="https://github.com/JuliaStats/StatsModels.jl/commit/9a5ba5cf9c6eac782494a6fe3eb09b772e2a64a5">9a5ba5cf9</a>">9a5ba5cf9</a><a href="https://github.com/JuliaStats/StatsModels.jl/commit/f1e2675632f2140aa891eaf5fc03ec8cc9bc92af">. Keep Nullable in cases where it's used to represent "software engineer's nulls", i.e. not missing values in data, since the replacement for that in Base is not yet ready. It should also be possible to avoid returning a Union{T, Missing} vector from predict() when the input variables do not allow for missing values by storing that in the type information, but this is left for later. Also require Julia 0.6, and remove useless call to float() which fails with the latest Nulls release.
311 of 330 relevant lines covered (94.24%)
95.55 hits per line
| ID | Job ID | Ran | Files | Coverage | |
|---|---|---|---|---|---|
| 1 | 154.1 | 0 |
94.24 |
Travis Job 154.1 | |
| 3 | 154.3 | 0 |
94.24 |
Travis Job 154.3 |