s_AddTree: Additive Tree: Tree-Structured Boosting C

In egenn/rtemis: Machine Learning and Visualization

Additive Tree: Tree-Structured Boosting C

Description

Train an Additive Tree model

Usage

s_AddTree(
  x,
  y = NULL,
  x.test = NULL,
  y.test = NULL,
  x.name = NULL,
  y.name = NULL,
  weights = NULL,
  update = c("exponential", "polynomial"),
  min.update = ifelse(update == "polynomial", 0.035, 1000),
  min.hessian = 0.001,
  min.membership = 1,
  steps.past.min.membership = 0,
  gamma = 0.8,
  max.depth = 30,
  learning.rate = 0.1,
  ifw = TRUE,
  ifw.type = 2,
  upsample = FALSE,
  downsample = FALSE,
  resample.seed = NULL,
  imetrics = TRUE,
  grid.resample.params = setup.resample("kfold", 5),
  metric = "Balanced Accuracy",
  maximize = TRUE,
  rpart.params = NULL,
  match.rules = TRUE,
  print.plot = FALSE,
  plot.fitted = NULL,
  plot.predicted = NULL,
  plot.theme = rtTheme,
  question = NULL,
  verbose = TRUE,
  prune.verbose = FALSE,
  trace = 1,
  grid.verbose = verbose,
  outdir = NULL,
  save.rpart = FALSE,
  save.mod = ifelse(!is.null(outdir), TRUE, FALSE),
  n.cores = rtCores
)

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
x.name	Character: Name for feature set
y.name	Character: Name for 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.
update	Character: "exponential" or "polynomial". Type of weight update. Default = "exponential"
min.update	Float: Minimum update for gradient step
min.hessian	[gS] Float: Minimum second derivative to continue splitting. Default = .001
min.membership	Integer: Minimum number of cases in a node. Default = 1
steps.past.min.membership	Integer: N steps to make past min.membership - For testing. Default = 0
gamma	[gS] Float: acceleration factor = lambda/(1 + lambda). Default = .8
max.depth	[gS] Integer: maximum depth of the tree. Default = 30
learning.rate	[gS] learning rate for the Newton Raphson step that updates the function values of the node
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)
imetrics	Logical: If TRUE, save interpretability metrics, i.e. N total nodes in tree and depth, in output. Default = TRUE
grid.resample.params	List: Output of setup.resample defining grid search parameters.
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.
rpart.params	List: rpart parameters, passed to rpart::rpart("parms")
match.rules	Logical: If TRUE, match cases to rules to get statistics per node, i.e. what percent of cases match each rule. If available, these are used by dplot3_addtree when plotting. Default = TRUE
print.plot	Logical: if TRUE, produce plot using mplot3 Takes precedence over plot.fitted and plot.predicted.
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
plot.theme	Character: "zero", "dark", "box", "darkbox"
question	Character: the question you are attempting to answer with this model, in plain language.
verbose	Logical: If TRUE, print summary to screen.
prune.verbose	Logical: If TRUE, prune tree.
trace	Integer: 0, 1, 2. The higher the number, the more verbose the output.
grid.verbose	Logical: Passed to gridSearchLearn
outdir	Path to output directory. If defined, will save Predicted vs. True plot, if available, as well as full model output, if save.mod is TRUE
save.rpart	Logical: passed to addtree
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)
n.cores	Integer: Number of cores to use.

Details

This function is for binary classification. The outcome must be a factor with two levels, the first level is the 'positive' class. Ensure there are no missing values in the data and that variables are either numeric (including integers) or factors. Use preprocess as needed to impute and convert characters to factors.

Factor levels should not contain the "/" character (it is used to separate conditions in the addtree object)

[gS] Indicates that more than one value can be supplied, which will result in grid search using internal resampling lambda = gamma/(1 - gamma)

Value

Object of class rtMod

Author(s)

E.D. Gennatas

References

Jose Marcio Luna, Efstathios D Gennatas, Lyle H Ungar, Eric Eaton, Eric S Diffenderfer, Shane T Jensen, Charles B Simone, Jerome H Friedman, Timothy D Solberg, Gilmer Valdes Building more accurate decision trees with the additive tree Proc Natl Acad Sci U S A. 2019 Oct 1;116(40):19887-19893. doi: 10.1073/pnas.1816748116

See Also

Other Supervised Learning: s_AdaBoost(), s_BART(), s_BRUTO(), s_BayesGLM(), s_C50(), s_CART(), s_CTree(), s_EVTree(), s_GAM(), s_GBM(), s_GLM(), s_GLMNET(), s_GLMTree(), s_GLS(), s_H2ODL(), s_H2OGBM(), s_H2ORF(), s_HAL(), s_Isotonic(), 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_BART(), s_C50(), s_CART(), s_CTree(), s_EVTree(), s_GBM(), 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 Interpretable models: s_C50(), s_CART(), s_GLM(), s_GLMNET(), s_GLMTree(), s_LMTree()

egenn/rtemis documentation built on Dec. 17, 2024, 6:16 p.m.