s_GBM: Gradient Boosting Machine [C, R, S]

In egenn/rtemis: Machine Learning and Visualization

Gradient Boosting Machine [C, R, S]

Description

Train a GBM model using gbm::gbm.fit

Usage

s_GBM(
  x,
  y = NULL,
  x.test = NULL,
  y.test = NULL,
  weights = NULL,
  ifw = TRUE,
  ifw.type = 2,
  upsample = FALSE,
  downsample = FALSE,
  resample.seed = NULL,
  distribution = NULL,
  interaction.depth = 2,
  shrinkage = 0.01,
  bag.fraction = 0.9,
  n.minobsinnode = 5,
  n.trees = 2000,
  max.trees = 5000,
  force.n.trees = NULL,
  gbm.select.smooth = FALSE,
  n.new.trees = 500,
  min.trees = 50,
  failsafe.trees = 500,
  imetrics = FALSE,
  .gs = FALSE,
  grid.resample.params = setup.resample("kfold", 5),
  gridsearch.type = "exhaustive",
  metric = NULL,
  maximize = NULL,
  plot.tune.error = FALSE,
  n.cores = rtCores,
  relInf = TRUE,
  varImp = FALSE,
  offset = NULL,
  var.monotone = NULL,
  keep.data = TRUE,
  var.names = NULL,
  response.name = "y",
  checkmods = FALSE,
  group = NULL,
  plot.perf = FALSE,
  plot.res = ifelse(!is.null(outdir), TRUE, FALSE),
  plot.fitted = NULL,
  plot.predicted = NULL,
  print.plot = FALSE,
  plot.theme = rtTheme,
  x.name = NULL,
  y.name = NULL,
  question = NULL,
  verbose = TRUE,
  trace = 0,
  grid.verbose = verbose,
  gbm.fit.verbose = FALSE,
  outdir = NULL,
  save.gridrun = FALSE,
  save.res = FALSE,
  save.res.mod = FALSE,
  save.mod = ifelse(!is.null(outdir), TRUE, FALSE)
)

Arguments

x	Numeric vector or matrix / data frame of features i.e. independent variables
y	Numeric vector of outcome, i.e. dependent variable
x.test	Numeric vector or matrix / data frame of testing set features Columns must correspond to columns in x
y.test	Numeric vector of testing set outcome
weights	Numeric vector: Weights for cases. For classification, weights takes precedence over ifw, therefore set weights = NULL if using ifw. Note: If weight are provided, ifw is not used. Leave NULL if setting ifw = TRUE.
ifw	Logical: If TRUE, apply inverse frequency weighting (for Classification only). Note: If weights are provided, ifw is not used.
ifw.type	Integer 0, 1, 2 1: class.weights as in 0, divided by min(class.weights) 2: class.weights as in 0, divided by max(class.weights)
upsample	Logical: If TRUE, upsample cases to balance outcome classes (for Classification only) Note: upsample will randomly sample with replacement if the length of the majority class is more than double the length of the class you are upsampling, thereby introducing randomness
downsample	Logical: If TRUE, downsample majority class to match size of minority class
resample.seed	Integer: If provided, will be used to set the seed during upsampling. Default = NULL (random seed)
distribution	Character: Distribution of the response variable. See gbm::gbm
interaction.depth	[gS] Integer: Interaction depth.
shrinkage	[gS] Float: Shrinkage (learning rate).
bag.fraction	[gS] Float (0, 1): Fraction of cases to use to train each tree. Helps avoid overfitting.
n.minobsinnode	[gS] Integer: Minimum number of observation allowed in node.
n.trees	Integer: Initial number of trees to fit
max.trees	Integer: Maximum number of trees to fit
force.n.trees	Integer: If specified, use this number of trees instead of tuning number of trees
gbm.select.smooth	Logical: If TRUE, smooth the validation error curve.
n.new.trees	Integer: Number of new trees to train if stopping criteria have not been met.
min.trees	Integer: Minimum number of trees to fit.
failsafe.trees	Integer: If tuning fails to find n.trees, use this number instead.
imetrics	Logical: If TRUE, save extra$imetrics with n.trees, depth, and n.nodes.
.gs	Internal use only
grid.resample.params	List: Output of setup.resample defining grid search parameters.
gridsearch.type	Character: Type of grid search to perform: "exhaustive" or "randomized".
metric	Character: Metric to minimize, or maximize if maximize = TRUE during grid search. Default = NULL, which results in "Balanced Accuracy" for Classification, "MSE" for Regression, and "Coherence" for Survival Analysis.
maximize	Logical: If TRUE, metric will be maximized if grid search is run.
plot.tune.error	Logical: If TRUE, plot the tuning error curve.
n.cores	Integer: Number of cores to use.
relInf	Logical: If TRUE (Default), estimate variables' relative influence.
varImp	Logical: If TRUE, estimate variable importance by permutation (as in random forests; noted as experimental in gbm). Takes longer than (default) relative influence. The two measures are highly correlated.
offset	Numeric vector of offset values, passed to gbm::gbm.fit
var.monotone	Integer vector with values 0, 1, -1 and length = N features. Used to define monotonicity constraints. 0: no constraint, 1: increasing, -1: decreasing.
plot.fitted	Logical: if TRUE, plot True (y) vs Fitted
plot.predicted	Logical: if TRUE, plot True (y.test) vs Predicted. Requires x.test and y.test
print.plot	Logical: if TRUE, produce plot using mplot3 Takes precedence over plot.fitted and plot.predicted.
plot.theme	Character: "zero", "dark", "box", "darkbox"
x.name	Character: Name for feature set
y.name	Character: Name for outcome
question	Character: the question you are attempting to answer with this model, in plain language.
verbose	Logical: If TRUE, print summary to screen.
grid.verbose	Logical: Passed to gridSearchLearn
outdir	Character: If defined, save log, 'plot.all' plots (see above) and RDS file of complete output
save.gridrun	Logical: If TRUE, save grid search models.
save.res.mod	Logical: If TRUE, save gbm model for each grid run. For diagnostic purposes only: Object size adds up quickly
save.mod	Logical: If TRUE, save all output to an RDS file in outdir save.mod is TRUE by default if an outdir is defined. If set to TRUE, and no outdir is defined, outdir defaults to paste0("./s.", mod.name)

Details

Early stopping is implemented by fitting n.trees initially, checking the optionally smoothed validation error curve, and adding n.new.trees if needed, until error does not reduce or max.trees is reached. [gS] in the argument description indicates that a vector of values can be passed, in which case grid search will be performed automatically using the resampling scheme defined by grid.resample.params.

This function includes a workaround for when gbm.fit fails. If an error is detected, gbm.fit is rerun until successful and the procedure continues normally

Author(s)

E.D. Gennatas

See Also

train_cv for external cross-validation

Other Supervised Learning: s_AdaBoost(), s_AddTree(), s_BART(), s_BRUTO(), s_BayesGLM(), s_C50(), s_CART(), s_CTree(), s_EVTree(), s_GAM(), s_GLM(), s_GLMNET(), s_GLMTree(), s_GLS(), s_H2ODL(), s_H2OGBM(), s_H2ORF(), s_HAL(), s_KNN(), s_LDA(), s_LM(), s_LMTree(), s_LightCART(), s_LightGBM(), s_MARS(), s_MLRF(), s_NBayes(), s_NLA(), s_NLS(), s_NW(), s_PPR(), s_PolyMARS(), s_QDA(), s_QRNN(), s_RF(), s_RFSRC(), s_Ranger(), s_SDA(), s_SGD(), s_SPLS(), s_SVM(), s_TFN(), s_XGBoost(), s_XRF()

Other Tree-based methods: s_AdaBoost(), s_AddTree(), s_BART(), s_C50(), s_CART(), s_CTree(), s_EVTree(), s_GLMTree(), s_H2OGBM(), s_H2ORF(), s_LMTree(), s_LightCART(), s_LightGBM(), s_MLRF(), s_RF(), s_RFSRC(), s_Ranger(), s_XGBoost(), s_XRF()

Other Ensembles: s_AdaBoost(), s_RF(), s_Ranger()

egenn/rtemis documentation built on Oct. 28, 2024, 6:30 a.m.