Single Layer Neural Network.
Calls nnet::nnet.formula()
from package nnet.
Note that modern neural networks with multiple layers are connected via package mlr3keras.
This Learner can be instantiated via the dictionary mlr_learners or with the associated sugar function lrn()
:
mlr_learners$get("classif.nnet") lrn("classif.nnet")
size
:
Adjusted default: 3L
Reason for change: no default in nnet()
.
Ripley BD (1996). Pattern Recognition and Neural Networks. Cambridge University Press. doi: 10.1017/cbo9780511812651 , https://doi.org/10.1017/cbo9780511812651.
mlr3::Learner
-> mlr3::LearnerClassif
-> LearnerClassifNnet
new()
Creates a new instance of this R6 class.
LearnerClassifNnet$new()
clone()
The objects of this class are cloneable with this method.
LearnerClassifNnet$clone(deep = FALSE)
deep
Whether to make a deep clone.
if (requireNamespace("nnet")) { learner = mlr3::lrn("classif.nnet") print(learner) # available parameters: learner$param_set$ids() }#> <LearnerClassifNnet:classif.nnet> #> * Model: - #> * Parameters: size=3 #> * Packages: nnet #> * Predict Type: prob #> * Feature types: numeric, factor, ordered #> * Properties: multiclass, twoclass, weights#> [1] "size" "subset" "na.action" "contrasts" "Wts" "mask" #> [7] "linout" "entropy" "softmax" "censored" "skip" "rang" #> [13] "decay" "maxit" "Hess" "trace" "MaxNWts" "abstol" #> [19] "reltol"