Constructs a learner class object for fitting a superlearner.
Usage
learner_sl(
learners,
info = NULL,
nfolds = 5L,
meta.learner = metalearner_nnls,
model.score = mse,
learner.args = NULL,
...
)Arguments
- learners
(list) List of learner objects (i.e. learner_glm)
- info
(character) Optional information to describe the instantiated learner object.
- nfolds
(integer) Number of folds to use in cross-validation to estimate the ensemble weights.
- meta.learner
(function) Algorithm to learn the ensemble weights (default non-negative least squares). Must be a function of the response (nx1 vector),
y, and the predictions (nxp matrix),pred, with p being the number of learners. Alternatively, this can be set to the character value "discrete", in which case the Discrete Super-Learner is applied where the model with the lowest risk (model-score) is given weight 1 and all other learners weight 0.- model.score
(function) Model scoring method (see learner)
- learner.args
(list) Additional arguments to learner$new().
- ...
Additional arguments to superlearner
Value
learner object.
Examples
sim1 <- function(n = 5e2) {
x1 <- rnorm(n, sd = 2)
x2 <- rnorm(n)
y <- x1 + cos(x1) + rnorm(n, sd = 0.5**.5)
data.frame(y, x1, x2)
}
d <- sim1()
m <- list(
"mean" = learner_glm(y ~ 1),
"glm" = learner_glm(y ~ x1 + x2),
"iso" = learner_isoreg(y ~ x1)
)
s <- learner_sl(m, nfolds = 10)
s$estimate(d)
pr <- s$predict(d)
if (interactive()) {
plot(y ~ x1, data = d)
points(d$x1, pr, col = 2, cex = 0.5)
lines(cos(x1) + x1 ~ x1, data = d[order(d$x1), ],
lwd = 4, col = lava::Col("darkblue", 0.3))
}
print(s)
#> ────────── learner object ──────────
#> superlearner
#> mean
#> glm
#> iso
#>
#> Estimate arguments: learners=<list>, nfolds=10, meta.learner=<function>, model.score=<function>
#> Predict arguments:
#> Formula: y ~ 1 <environment: 0x561eb2233d30>
#> ─────────────────────────────────────
#> score weight
#> mean 4.8684025 0.09195207
#> glm 0.8579904 0.07689746
#> iso 0.4668761 0.83115047
# weights(s$fit)
# score(s$fit)
cvres <- cv(s, data = d, nfolds = 3, rep = 2)
cvres
#>
#> 3-fold cross-validation with 2 repetitions
#>
#> ── mse
#> mean sd min max
#> sl 0.49724 0.04312 0.45503 0.55521
#> mean 4.86631 0.29296 4.55459 5.23194
#> glm 0.86610 0.05665 0.76497 0.91749
#> iso 0.47305 0.01645 0.44241 0.49028
#>
#> ── mae
#> mean sd min max
#> sl 0.56817 0.02602 0.54496 0.61206
#> mean 1.76295 0.06099 1.66893 1.85225
#> glm 0.75561 0.03516 0.69341 0.79557
#> iso 0.55661 0.01008 0.54620 0.56973
#>
#> ── weight
#> mean sd min max
#> sl - - - -
#> mean 0.07748 0.04056 0.02455 0.11796
#> glm 0.09614 0.01524 0.07534 0.12235
#> iso 0.82638 0.04238 0.78781 0.87947
# coef(cvres)
# score(cvres)