pp.settings-class: (Internal) A reference class to keep settings/information of...

In Marchen/partial.plot: Partial dependence plot for statistical/machine learning models.

(Internal) A reference class to keep settings/information of partial.plot.

Description

This reference class keeps information used for partial.plot. This class is not intended to be used by users directly.

Fields

adapter

an instance of model.adapter.

model

a supported model object used for plotting.

parman

an object of par.manager class.

x.names

a character vector representing names of explanatory variables used for plotting.

x.names.factor

a character vector representing names of factor explanatory variables used for plotting.

x.names.numeric

a character vector representing names of numeric explanatory variables used for plotting.

type

a character literal indicating type of scale for plotting. This is similar to type argument of many predict methods. Possible values are "response", "link" and "prob". If "link" is specified, partial relationship and residuals are drawn in the scale of the linear predictor. On the other hand, partial relationship and residuals are drawn in the scale of the response variable if "response" is specified. For classification models, only "prob", which calculate probability of a specific class, can be used.

positive.class

a class for which predicted probability is calculated. If not specified, first class of the factor or first unique value of the response variable is used.

plot.type

a character literal to indicate plot type. Possible values are "2D" and "3D".

factor.levels

a list having levels of factors specified by x.names.

group

a vector of grouping levels.

unique.group

a vector of unique grouping levels.

numeric.sequences

a list having sequences of numeric variables specified by x.names.

data

a data.frame containing data used for plotting.

fun.3d

the function draws 3D graphs. Possibly persp and images can work.

draw.residual

a logical representing that partial.plot draws residual points.

draw.relationship

a logical representing that partial.plot draws predicted relationships.

draw.interval

a logical representing that partial.plot draws intervals of predicted relationships.

draw.hist

a logical specifying whether a histogram should be added to the graph.

interval.levels

numeric vector indicating confidence level or quantiles for predicted relationships.

resolution

an integer specifying resolution of lines, polygons, wireframes, and images of numeric variables.

xlab

label of X axis.

ylab

label of Y axis.

zlab

label of Z axis.

add

logical indicating whether graphic elements are added to existing plot.

sep

a character representing separator of grouping factor levels.

extrapolate

a logical indicating whether extrapolation is allowed for predicted relationships.

other.pars

a list containing other graphic parameters passed to partial.plot().

n.cores

an integer specifing number of processes used for multiprocessing.

relationship

a data.frame having partial relationship data.

relationship.split

a list having partial relationship data splitted for each group.

residual

a numeric vector having partial residual data.

has.relationship

a logical indicating the object has partial relationship data.

has.residual

a logical indicating the object has partial residual data.

Methods

check.data.access()

Check availability of data.

check.intervals()

Check interval of predicted relationships.

check.labels()

Test correctness of xlab and ylab.

check.params()

Check errors in specified parameters and stop if any errors are found.

check.positive.class()

Check whether specified positive class is in response variable.

check.resolution()

Check resolution is integer.

check.type()

Check specified type.

check.x.names()

Check correctness of x.names.

cluster.apply(X, FUN, ...)

Interface for lapply() or clusterApply()

create.numeric.sequence(x.name)

Create sequence of an explanatory variable based on original range of data.

init.factor.levels()

Initialize factor levels.

init.group()

Initialize a vector representing data splitting group.

init.intervals()

Initialize settings of intervals.

init.labels()

Set xlab, ylab and zlab for 2D & 3D plot.

init.multiprocessing()

Initialize multiprocessing.

init.numeric.sequences()

Initialize numeric sequences.

init.numeric.sequences.2d()

Initialize numeric sequences.

init.numeric.sequences.3d()

Initialize numeric sequences.

init.plot.type()

Determine plot type (2D/3D).

init.positive.class()

Initialize positive class.

init.resolution()

Initialize resolution

init.x.names()

Initialize names of factor and numeric explanatory variables.

initialize( model, x.names, data = NULL, type = "response", positive.class = "", fun.3d = persp, draw.residual = TRUE, draw.relationship = TRUE, draw.interval = TRUE, draw.hist = FALSE, interval.levels = 0.95, resolution = NULL, xlab = NULL, ylab = NULL, zlab = NULL, add = FALSE, sep = " - ", extrapolate = FALSE, n.cores = NULL, ... )

Initialize pp.settings object.

model

a supported model object used for plotting.

x.names

a character vector representing names of explanatory variables used for plotting.

data

a data.frame containing data used for plotting.

type

type of relationship to draw. Possible values are 'response', 'link' and 'prob'.

positive.class

class label for which probability is calculated. By default, partial.plot calculate probabilities for first class of the factor.

fun.3d

the function used for drawing 3D relationship graphs. Possibly, persp and image can work.

draw.residual

whether the partial.plot draws residual points.

draw.relationship

whether partial.plot draws predicted relationships.

draw.interval

whetehr partial.plot draws interval of predicted relationships.

draw.hist

whetehr partial.plot draws a histogram of explanatory variable.

interval.levels

numeric vector specifying level of confidence intervals/quantile.

resolution

an integer specifying resolution of lines, polygons, wireframes, and images of numeric variables.

xlab, ylab, zlab

a character specifying used for label of X/Y/Z axis.

add

logical specifying whether graphic elements are added to the existing plot.

sep

a character representing separator of grouping factor levels.

exprapolate

a logical indicating whether extrapolation is allowed for predicted relationships.

n.cores

an integer representing number of processes used for calculation. If NULL is specified, maximum number of logical processors are used. This value is ignored when the models compatible with lsmeans are specified.

...

other graphic parameters.

potential.arg.names(fun)

Return a character vector having names of potential arguments of the function.

fun

function to get potential argument names.

set.function.args(function.args = list(), fun = .self$fun.3d)

Assign function arguments kept in this class and returns it in a list.

function.args

arguments for the function call to be modified.

fun

target function to be call.

set.relationship(object)

Set partial relationship data.

set.residual(object)

Import partial residual data.

update.pars( fun.3d, draw.residual, draw.relationship, draw.interval, draw.hist, xlab, ylab, zlab, col, lty, lwd, pch, ... )

Update settings of (mainly) graphic parameters for data reusing. fun.3d, draw.residual, draw.relationship, draw.interval, xlab, ylab, zlab, col, lty, lwd, pch and ... can be set.

Marchen/partial.plot documentation built on Feb. 13, 2025, 4:18 a.m.