Skip to content

Commit 7e16174

Browse files
committed
added .travis.yml file
1 parent fcaf1d5 commit 7e16174

File tree

2 files changed

+29
-1
lines changed

2 files changed

+29
-1
lines changed

.travis.yml

Lines changed: 27 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,27 @@
1+
language: r
2+
3+
cache: packages
4+
5+
sudo: false
6+
7+
r:
8+
9+
- oldrel
10+
11+
- release
12+
13+
- devel
14+
15+
addons:
16+
17+
apt:
18+
19+
packages: libmpfr-dev
20+
21+
warnings_are_errors: true
22+
23+
before_install:
24+
25+
- tlmgr install index # for texlive and vignette?
26+
27+
- cd selectiveInference

selectiveInference/man/fixedLassoInf.Rd

Lines changed: 2 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -10,7 +10,7 @@ fixed value of the tuning parameter lambda
1010
}
1111
\usage{
1212
fixedLassoInf(x, y, beta, lambda, family = c("gaussian", "binomial",
13-
"cox"),intercept=TRUE, status=NULL, sigma=NULL, alpha=0.1,
13+
"cox"),intercept=TRUE, add.targets=NULL, status=NULL, sigma=NULL, alpha=0.1,
1414
type=c("partial","full"), tol.beta=1e-5, tol.kkt=0.1,
1515
gridrange=c(-100,100), bits=NULL, verbose=FALSE)
1616
}
@@ -56,6 +56,7 @@ Significance level for confidence intervals (target is miscoverage alpha/2 in ea
5656
Was the lasso problem solved (e.g., by glmnet) with an intercept in the model?
5757
Default is TRUE. Must be TRUE for "binomial" family. Not used for 'cox" family, where no intercept is assumed.
5858
}
59+
\item{add.targets}{Optional vector of predictors to be included as targets of inference, regardless of whether or not they are selected by the lasso. Default is NULL.}
5960
\item{status}{Censoring status for Cox model; 1=failurem 0=censored}
6061
\item{type}{Contrast type for p-values and confidence intervals: default is
6162
"partial"---meaning that the contrasts tested are the partial population

0 commit comments

Comments
 (0)