ck37r (version 1.0.3)

cvsl_weights: Create a table of meta-weights from a CV.SuperLearner

Description

Returns summary statistics (mean, sd, min, max) on the distribution of the weights assigned to each learner across SuperLearner ensembles. This makes it easier to understand the stochastic nature of the SL learner weights and to see how often certain learners are used. This function may eventually be moved into the SuperLearner package.

Usage

cvsl_weights(
  cvsl,
  sort = T,
  nonzero = F,
  clean_names = T,
  rank = T,
  digits = 5
)

Arguments

cvsl

CV.SuperLearner result object

sort

If TRUE sort rows (learners) in descending order by mean weight.

nonzero

Restrict to learners with a non-zero mean weight.

clean_names

Remove "SL." from the front and "_All" from the end of learner names.

rank

Adding the learner rank to the table.

digits

Number of digits to round the results. Set to NULL to disable.

Value

Table in data frame form with each learner's mean, sd, min, and max meta-weight in the ensemble of each learner.

References

Polley EC, van der Laan MJ (2010) Super Learner in Prediction. U.C. Berkeley Division of Biostatistics Working Paper Series. Paper 226. http://biostats.bepress.com/ucbbiostat/paper266/

van der Laan, M. J., Polley, E. C. and Hubbard, A. E. (2007) Super Learner. Statistical Applications of Genetics and Molecular Biology, 6, article 25. http://www.degruyter.com/view/j/sagmb.2007.6.issue-1/sagmb.2007.6.1.1309/sagmb.2007.6.1.1309.xml

Examples

# NOT RUN {
library(SuperLearner)
library(ck37r)

data(Boston, package = "MASS")

set.seed(1, "L'Ecuyer-CMRG")

# Subset rows to speed up example computation.
row_subset = sample(nrow(Boston), 100)

Boston = Boston[row_subset, ]
X = subset(Boston, select = -chas)

cvsl = CV.SuperLearner(Boston$chas, X[, 1:2], family = binomial(),
                      cvControl = list(V = 2, stratifyCV = TRUE),
                      SL.library = c("SL.mean", "SL.glm"))
cvsl_weights(cvsl)

# }