learners: List of supported learning algorithms.
Description
{
Boosting from ada
package: ada}
- classif.boosting
{
Boosting from adabag
package: boosting
Note that
xval has been set to 0 by default for speed.}
- classif.blackboost
{
Gradient boosting
with regression trees from mboost package:
blackboost}
- classif.ctree
{
Conditional Inference
Trees from party package: ctree}
- classif.fnn
{
Fast k-Nearest Neighbor
from FNN package: knn}
- classif.gbm
{
Gradient boosting machine
from gbm package: gbm}
- classif.glmboost
{
Boosting for GLMs from
mbboost package: glmboost
Note
that family has been set to Binomial() by
default.} - classif.J48
{
J48 Decision
Trees from RWeka package: J48} Note
that NAs are directly passed to WEKA with na.action
= na.pass. - classif.JRip
{
Propositional
Rule Learner from RWeka package:
JRip} Note that NAs are directly
passed to WEKA with na.action = na.pass.
- classif.kknn
{
k-Nearest Neighbor from
kknn package: kknn}
- classif.ksvm
{
Support Vector Machines
from kernlab package: ksvm
Note
that kernel parameters have to be passed directly and not
by using the kpar list in ksvm.
Note that fit
has been set to FALSE by default for speed.}
- classif.lda
{
Linear Discriminant
Analysis from MASS package: lda}
- classif.logreg
{
Logistic Regression from
stats package: glm}
- classif.lssvm
{
Least Squares Support
Vector Machine from kernlab package:
lssvm
Note that fitted
has been set to FALSE by default for speed.}
- classif.lvq1
{
Learning Vector
Quantization from class package:
lvq1} - classif.mda
{
Mixture Discriminant Analysis from mda package:
mda
Note that keep.fitted
has been set to FALSE by default for speed.}
- classif.multinom
{
Multinomial Regression
from nnet package: multinom}
- classif.naiveBayes
{
Naive Bayes from
e1071 package: naiveBayes}
- classif.nnet
{
Neural Network from nnet
package: nnet
Note that
size has been set to 3 by default.}
- classif.OneR
{
1-R classifier from RWeka
package: OneR} Note that NAs are
directly passed to WEKA with na.action = na.pass.
- classif.PART
{
PART decision lists from
RWeka package: PART} Note that NAs
are directly passed to WEKA with na.action =
na.pass. - classif.qda
{
Quadratic
Discriminant Analysis from MASS package:
qda}
- classif.randomForest
{
Random Forest from
randomForest package:
randomForest}
- classif.rda
{
Regularized Discriminant
Analysis from klaR package: rda
Note that estimate.error has been set to
FALSE by default for speed.}
- classif.rpart
{
Decision Tree from rpart
package: rpart
Note that
xval has been set to 0 by default for speed.}
- classif.svm
{
Support Vector Machines
(libsvm) from e1071 package: svm}Details
{
Gradient
boosting with regression trees from mboost package:
blackboost}
- regr.earth
{
Multivariate Adaptive
Regression Splines from earth package:
earth} - regr.fnn
{
Fast k-Nearest Neighbor from FNN package:
knn} - regr.gbm
{
Gradient boosting machine from gbm package:
gbm
Note that distribution
has been set to gaussian by default.}
- regr.kknn
{
K-Nearest-Neighbor regression
from kknn package: kknn}
- regr.km
{
Kriging from DiceKriging
package: km}
- regr.ksvm
{
Support Vector Machines from
kernlab package: ksvm
Note that
kernel parameters have to be passed directly and not by
using the kpar list in ksvm.
Note that fit has
been set to FALSE by default for speed.}
- regr.penalized.lasso
{
Lasso regression
from penalized package:
penalized}
- regr.lm
{
Simple linear regression from
stats package: lm}
- regr.mars
{
Multivariate Adaptive
Regression Splines from mda package:
mars} - regr.nnet
{
Neural Network from nnet package:
nnet
Note that size has
been set to 3 by default.} - regr.pcr
{
Principal component regression from pls package:
pcr
Note that model has been
set to FALSE by default for speed.}
- regr.randomForest
{
Random Forest from
randomForest package:
randomForest}
- regr.penalized.ridge
{
Ridge regression
from penalized package:
penalized}
- regr.rpart
{
Decision Tree from rpart
package: rpart
Note that
xval has been set to 0 by default for speed.}
- regr.rsm
{
Response surface regression
from rsm package: rsm
Note that you
select the order of the regression by using modelfun="FO"
(first order), "TWI" (two-way interactions, this is with
1st oder terms!) and "SO" (full second order)}.
- regr.rvm
{
Relevance Vector Machine from
rpart kernlab: rvm
Note that
kernel parameters have to be passed directly and not by
using the kpar list in rvm.
Note that fit has
been set to FALSE by default for speed.}