Type: Package
Title: Individual Conditional Expectation Plot Toolbox
Version: 1.1.5
Date: 2022-08-18
Author: Alex Goldstein, Adam Kapelner, Justin Bleich
Maintainer: Adam Kapelner <kapelner@qc.cuny.edu>
Description: Implements Individual Conditional Expectation (ICE) plots, a tool for visualizing the model estimated by any supervised learning algorithm. ICE plots refine Friedman's partial dependence plot by graphing the functional relationship between the predicted response and a covariate of interest for individual observations. Specifically, ICE plots highlight the variation in the fitted values across the range of a covariate of interest, suggesting where and to what extent they may exist.
License: GPL-2 | GPL-3
Depends: sfsmisc
Suggests: randomForest, MASS
NeedsCompilation: no
Packaged: 2022-08-22 14:00:12 UTC; kapel
Repository: CRAN
Date/Publication: 2022-08-22 14:20:10 UTC

Clustering of ICE and d-ICE curves by kmeans.

Description

Clustering if ICE and d-ICE curves by kmeans. All curves are centered to have mean 0 and then kmeans is applied to the curves with the specified number of clusters.

Usage

clusterICE(ice_obj, nClusters, plot = TRUE, plot_margin = 0.05, 
           colorvec, plot_pdp = FALSE, x_quantile = FALSE, 
           avg_lwd = 3, centered = FALSE, 
           plot_legend = FALSE, ...)

Arguments

ice_obj

Object of class ice or dice to cluster.

nClusters

Number of clusters to find.

plot

If TRUE, plots the clusters.

plot_margin

Extra margin to pass to ylim as a fraction of the range of cluster centers.

colorvec

Optional vector of colors to use for each cluster.

plot_pdp

If TRUE, the PDP (ice object) or d-PDP (dice object) is plotted with a dotted black line and highlighted in yellow.

x_quantile

If TRUE, the plot is drawn with the x-axis taken to be quantile(gridpts). If FALSE, the predictor's original scale is used.

avg_lwd

Average line width to use when plotting the cluster means. Line width is proportional to the cluster's size.

centered

If TRUE, all cluster means are shifted to be to be 0 at the minimum value of the predictor. If FALSE, the original cluster means are used.

plot_legend

If TRUE a legend mapping line colors to the proportion of the data in each cluster is added to the plot.

...

Additional arguments for plotting.

Value

The ouput of the kmeans call (a list of class kmeans).

See Also

ice, dice

Examples

## Not run: 
require(ICEbox)
require(randomForest)
require(MASS) #has Boston Housing data, Pima

data(Boston) #Boston Housing data
X = Boston
y = X$medv
X$medv = NULL

## build a RF:
bh_rf = randomForest(X, y)

## Create an 'ice' object for the predictor "age":
bh.ice = ice(object = bh_rf, X = X, y = y, predictor = "age",
            frac_to_build = .1) 

## cluster the curves into 2 groups.
clusterICE(bh.ice, nClusters = 2, plot_legend = TRUE)

## cluster the curves into 3 groups, start all at 0.
clusterICE(bh.ice, nClusters = 3, plot_legend = TRUE, center = TRUE)

## End(Not run)

Creates an object of class dice.

Description

Estimates the partial derivative function for each curve in an ice object. See Goldstein et al (2013) for further details.

Usage

dice(ice_obj, DerivEstimator)

Arguments

ice_obj

Object of class ice. This function generates partial derivative estimates for each row in ice_obj$ice_curves.

DerivEstimator

Optional function with a single argument y. Returns the estimated partial derivative of a function sampled at the points (ice_obj$gridpts,y). If omitted, the default (a) smooths (ice_obj$gridpts,y) using supsmu and then (b) uses the D1tr function ("discrete first derivative using simple difference ratios") found in the sfsmisc package to estimate the derivative.

Value

A list of class dice with the following elements. Most are passed directly through from ice_object and exist to enable various plotting facilities.

d_ice_curves

Matrix of dimension nrow(Xice) by length(gridpts). Each row corresponds to an observation's d-ICE curve, estimated at the values of predictor in gridpts.

xj

The actual values of predictor observed in the data in the order of Xice.

actual_deriv

Vector of length nrow(Xice) containing the estimated partial derivatives at the value of the predictor actually found in Xice.

sd_deriv

Vector of length length(gridpts) with the cross-observation sd of partial derivative estimates. For instance sd_deriv[1] equals sd(d_ice_curves[,1]).

logodds

Passed from ice_object. If TRUE, d_ice_curves are estimated derivatives of the centered log-odds.

gridpts

Passed from ice_object.

predictor

Passed from ice_object.

xlab

Passed from ice_object.

nominal_axis

Passed from ice_object.

range_y

Passed from ice_object.

Xice

Passed from ice_object.

dpdp

The estimated partial derivative of the PDP.

References

Goldstein, A., Kapelner, A., Bleich, J., and Pitkin, E., Peeking Inside the Black Box: Visualizing Statistical Learning With Plots of Individual Conditional Expectation. (2014) Journal of Computational and Graphical Statistics, in press

Martin Maechler et al. sfsmisc: Utilities from Seminar fuer Statistik ETH Zurich. R package version 1.0-24.

See Also

plot.dice, print.dice, summary.dice

Examples

 ## Not run: 
# same examples as for 'ice', but now create a derivative estimate as well.
require(ICEbox)
require(randomForest)
require(MASS) #has Boston Housing data, Pima

########  regression example
data(Boston) #Boston Housing data
X = Boston
y = X$medv
X$medv = NULL

## build a RF:
bhd_rf_mod = randomForest(X, y)

## Create an 'ice' object for the predictor "age":
bhd.ice = ice(object = bhd_rf_mod, X = X, y = y, predictor = "age", frac_to_build = .1)

# make a dice object:
bhd.dice = dice(bhd.ice)

#### classification example
data(Pima.te)  #Pima Indians diabetes classification
y = Pima.te$type
X = Pima.te
X$type = NULL

## build a RF:
pima_rf = randomForest(x = X, y = y)

## Create an 'ice' object for the predictor "skin":
# For classification we plot the centered log-odds. If we pass a predict
# function that returns fitted probabilities, setting logodds = TRUE instructs
# the function to set each ice curve to the centered log-odds of the fitted 
# probability. 
pima.ice = ice(object = pima_rf, X = X, predictor = "skin", logodds = TRUE,
                    predictfcn = function(object, newdata){ 
                         predict(object, newdata, type = "prob")[, 2]
                    }
              )

# make a dice object:
pima.dice = dice(pima.ice)

## End(Not run)

Creates an object of class ice.

Description

Creates an ice object with individual conditional expectation curves for the passed model object, X matrix, predictor, and response. See Goldstein et al (2013) for further details.

Usage

ice(object, X, y, predictor, predictfcn, verbose = TRUE, frac_to_build = 1, 
             indices_to_build = NULL, num_grid_pts, logodds = FALSE, probit = FALSE, ...)

Arguments

object

The fitted model to estimate ICE curves for.

X

The design matrix we wish to estimate ICE curves for. Rows are observations, columns are predictors. Typically this is taken to be object's training data, but this is not strictly necessary.

y

Optional vector of the response values object was trained on. It is used to compute y-axis ranges that are useful for plotting. If not passed, the range of predicted values is used and a warning is printed.

predictor

The column number or variable name in X of the predictor of interest, (x_S = X[, j]).

predictfcn

Optional function that accepts two arguments, object and newdata, and returns an N vector of object's predicted response for data newdata. If this argument is not passed, the procedure attempts to find a generic predict function corresponding to class(object).

verbose

If TRUE, prints messages about the procedure's progress.

frac_to_build

Number between 0 and 1, with 1 as default. For large X matrices or fitted models that are slow to make predictions, specifying frac_to_build less than 1 will choose a subset of the observations to build curves for. The subset is chosen such that the remaining observations' values of predictor are evenly spaced throughout the quantiles of the full X[,predictor] vector.

indices_to_build

Vector of indices, \subset \{1, \ldots, nrow(X)\} specifying which observations to build ICE curves for. As this is an alternative to setting frac_to_build, both cannot be specified.

num_grid_pts

Optional number of values in the range of predictor at which to estimate each curve. If missing, the curves are estimated at each unique value of predictor in the X observations we estimate ICE curves for.

logodds

If TRUE, for classification creates PDPs by plotting the centered log-odds implied by the fitted probabilities. We assume that the generic or passed predict function returns probabilities, and so the flag tells us to transform these to centered logits after the predictions are generated. Note: probit cannot be TRUE.

probit

If TRUE, for classification creates PDPs by plotting the probit implied by the fitted probabilities. We assume that the generic or passed predict function returns probabilities, and so the flag tells us to transform these to probits after the predictions are generated. Note: logodds cannot be TRUE.

...

Other arguments to be passed to object's generic predict function.

Value

A list of class ice with the following elements.

gridpts

Sorted values of predictor at which each curve is estimated. Duplicates are removed – by definition, elements of gridpts are unique.

ice_curves

Matrix of dimension nrow(X) by length(gridpts). Each row corresponds to an observation's ICE curve, estimated at the values of predictor in gridpts.

xj

The actual values of predictor observed in the data in the order of Xice.

actual_predictions

Vector of length nrow(X) containing the model's predictions at the actual value of the predictors in the order of Xice.

xlab

String with the predictor name corresponding to predictor. If predictor is a column number, xlab is set to colnames(X)[, predictor].

nominal_axis

If TRUE, length(gridpts) is 5 or fewer; otherwise FALSE. When TRUE the plot function treats the x-axis as if x is nominal.

range_y

If y was passed, the range of the response. Otherwise it defaults to be
max(ice_curves) - min(ice_curves) and a message is printed to the console.

sd_y

If y was passed, the standard deviation of the response. Otherwise it is defaults to sd(actual_predictions) and a message is printed to the console.

Xice

A matrix containing the subset of X for which ICE curves are estimated. Observations are ordered to be increasing in predictor. This ordering is the same one as in ice_curves, xj and actual_predictions, meaning for all these objects the i-th element refers to the same observation in X.

pdp

A vector of size length(gridpts) which is a numerical approximation to the partial dependence function (PDP) corresponding to the estimated ICE curves. See Goldstein et al (2013) for a discussion of how the PDP is a form of post-processing. See Friedman (2001) for a description of PDPs.

predictor

Same as the argument, see argument description.

logodds

Same as the argument, see argument description.

indices_to_build

Same as the argument, see argument description.

frac_to_build

Same as the argument, see argument description.

predictfcn

Same as the argument, see argument description.

References

Jerome Friedman. Greedy Function Approximation: A Gradient Boosting Machine. The Annals of Statistics, 29(5): 1189-1232, 2001.

Goldstein, A., Kapelner, A., Bleich, J., and Pitkin, E., Peeking Inside the Black Box: Visualizing Statistical Learning With Plots of Individual Conditional Expectation. (2014) Journal of Computational and Graphical Statistics, in press

See Also

plot.ice, print.ice, summary.ice

Examples

## Not run: 
require(ICEbox)
require(randomForest)
require(MASS) #has Boston Housing data, Pima

########  regression example
data(Boston) #Boston Housing data
X = Boston
y = X$medv
X$medv = NULL

## build a RF:
bhd_rf_mod = randomForest(X, y)

## Create an 'ice' object for the predictor "age":
bhd.ice = ice(object = bhd_rf_mod, X = X, y = y, predictor = "age", frac_to_build = .1) 

#### classification example
data(Pima.te)  #Pima Indians diabetes classification
y = Pima.te$type
X = Pima.te
X$type = NULL

## build a RF:
pima_rf_mod = randomForest(x = X, y = y)

## Create an 'ice' object for the predictor "skin":
# For classification we plot the centered log-odds. If we pass a predict
# function that returns fitted probabilities, setting logodds = TRUE instructs
# the function to set each ice curve to the centered log-odds of the fitted 
# probability.
pima.ice = ice(object = pima_rf_mod, X = X, predictor = "skin", logodds = TRUE,
                    predictfcn = function(object, newdata){ 
                         predict(object, newdata, type = "prob")[, 2]
                    }
              )


## End(Not run)

Create a plot of a dice object.

Description

Plotting of dice objects.

Usage

## S3 method for class 'dice'
plot(x, plot_margin = 0.05, frac_to_plot = 1, 
      plot_sd = TRUE, plot_orig_pts_deriv = TRUE, pts_preds_size = 1.5, 
      colorvec, color_by = NULL, x_quantile = TRUE, plot_dpdp = TRUE, 
      rug_quantile = seq(from = 0, to = 1, by = 0.1), ...)

Arguments

x

Object of class dice to plot.

plot_margin

Extra margin to pass to ylim as a fraction of the range of x$d_ice_curves.

frac_to_plot

If frac_to_plot is less than 1, randomly plot frac_to_plot fraction of the curves in x$d_ice_curves.

plot_sd

If TRUE, plot the cross-observation sd of partial derivatives below the derivative plots.

plot_orig_pts_deriv

If TRUE, marks each curve at the location of the derivative estimate at the location of predictor actually occurring in the data. If FALSE no mark is drawn.

pts_preds_size

Size of points to make if plot_orig_pts_deriv is TRUE.

colorvec

Optional vector of colors to use for each curve.

color_by

Optional variable name (or column number) in Xice to color curves by. If the color_by variable has 10 or fewer unique values, a discrete set of colors is used for each value and a legend is printed and returned. If there are more values, curves are colored from light to dark corresponding to low to high values of the variable specified by color_by.

x_quantile

If TRUE, the plot is drawn with the x-axis taken to be quantile(gridpts). If FALSE, the predictor's original scale is used.

plot_dpdp

If TRUE, the estimated derivative of the PDP is plotted and highlighted in yellow.

rug_quantile

If not null, tick marks are drawn on the x-axis corresponding to the vector of quantiles specified by this parameter. Forced to NULL when x_quantile is set to TRUE.

...

Additional plotting arguments.

Value

A list with the following elements.

plot_points_indices

Row numbers of Xice of those observations presented in the plot.

legend_text

If the color_by argument was used, a legend describing the map between the color_by predictor and curve colors.

See Also

dice

Examples

## Not run: 
require(ICEbox)
require(randomForest)
require(MASS) #has Boston Housing data, Pima

data(Boston) #Boston Housing data
X = Boston
y = X$medv
X$medv = NULL

## build a RF:
bhd_rf_mod = randomForest(X, y)

## Create an 'ice' object for the predictor "age":
bhd.ice = ice(object = bhd_rf_mod, X = X, y = y, predictor = "age", frac_to_build = .1) 

# estimate derivatives, then plot.
bhd.dice = dice(bhd.ice)
plot(bhd.dice)

## End(Not run)

Plotting of ice objects.

Description

Plotting of ice objects.

Usage

## S3 method for class 'ice'
plot(x, plot_margin = 0.05, frac_to_plot = 1, 
	 plot_points_indices = NULL, plot_orig_pts_preds = TRUE, 
     pts_preds_size = 1.5, colorvec, color_by = NULL, 
     x_quantile = TRUE, plot_pdp = TRUE,
     centered = FALSE, prop_range_y = TRUE, 
     rug_quantile = seq(from = 0, to = 1, by = 0.1),
     centered_percentile = 0, 
     point_labels = NULL, point_labels_size = NULL, 
     prop_type,...)

Arguments

x

Object of class ice to plot.

plot_margin

Extra margin to pass to ylim as a fraction of the range of x$ice_curves.

frac_to_plot

If frac_to_plot is less than 1, randomly plot frac_to_plot fraction of the curves in x$ice_curves.

plot_points_indices

If not NULL, this plots only the indices of interest. If not NULL, frac_to_plot must be 1 otherwise an error is thrown. Default is NULL.

plot_orig_pts_preds

If TRUE, marks each curve at the location of the observation's actual fitted value. If FALSE, no mark is drawn.

pts_preds_size

Size of points to make if plot_origin_pts_preds is TRUE.

colorvec

Optional vector of colors to use for each curve.

color_by

Optional variable name in Xice, column number in Xice, or data vector of the correct length to color curves by. If the color_by variable has 10 or fewer unique values, a discrete set of colors is used for each value and a legend is printed and returned. If there are more values, curves are colored from light to dark corresponding to low to high values of the variable specified by color_by.

x_quantile

If TRUE, the plot is drawn with the x-axis taken to be quantile(gridpts). If FALSE, the predictor's original scale is used.

plot_pdp

If TRUE, the PDP is plotted and highlighted in yellow.

centered

If TRUE, all curves are re-centered to be 0 at the quantile given by
centered_percentile. See Goldstein et al (2013) for details and examples. If FALSE, the original ice_curves are plotted.

prop_range_y

When TRUE and centered=TRUE as well, the range of the right vertical axis displays the centered values as a fraction of the sd of the fitted values on actual observations if prop_type is missing or set to "sd". If prop_type is set to "range", the right axis displays the centered values as a fraction of the range of the fitted values over the actual observations.

centered_percentile

The percentile of predictor for which all ice_curves are "pinched together" and set to be 0. Default is .01.

point_labels

If not NULL, labels to plot next to each point. Default is NULL.

point_labels_size

If not NULL, size of labels to plot next to each point. Default is NULL which means it's the size of pts_preds_size.

rug_quantile

If not NULL, tick marks are drawn on the x-axis corresponding to the vector of quantiles specified by this parameter. Forced to NULL when x_quantile is set to TRUE.

prop_type

Scaling factor for the right vertical axis in centered plots if prop_range_y is TRUE. Can be one of "sd" (default) or "range". Ignored if centered and prop_range_y are not both TRUE.

...

Other arguments to be passed to the plot function.

Value

A list with the following elements.

plot_points_indices

Row numbers of Xice of those observations presented in the plot.

legend_text

If the color_by argument was used, a legend describing the map between the color_by predictor and curve colors.

See Also

ice

Examples

## Not run: 
require(ICEbox)
require(randomForest)
require(MASS) #has Boston Housing data, Pima

data(Boston) #Boston Housing data
X = Boston
y = X$medv
X$medv = NULL

## build a RF:
bhd_rf_mod = randomForest(X, y)

## Create an 'ice' object for the predictor "age":
bhd.ice = ice(object = bhd_rf_mod, X = X, y = y, predictor = "age",
            frac_to_build = .1) 

## plot
plot(bhd.ice, x_quantile = TRUE, plot_pdp = TRUE, frac_to_plot = 1) 

## centered plot
plot(bhd.ice, x_quantile = TRUE, plot_pdp = TRUE, frac_to_plot = 1, 
		centered = TRUE) 

## color the curves by high and low values of 'rm'.
# First create an indicator variable which is 1 if the number of 
# rooms is greater than the median:
median_rm = median(X$rm)
bhd.ice$Xice$I_rm = ifelse(bhd.ice$Xice$rm > median_rm, 1, 0)  

plot(bhd.ice, frac_to_plot = 1, centered = TRUE, prop_range_y = TRUE,  
            x_quantile = T, plot_orig_pts_preds = T, color_by = "I_rm")
bhd.ice = ice(object = bhd_rf_mod, X = X, y = y, predictor = "age",
            frac_to_build = 1)             
plot(bhd.ice, frac_to_plot = 1, centered = TRUE, prop_range_y = TRUE,  
            x_quantile = T, plot_orig_pts_preds = T, color_by = y)            

## End(Not run)

Print method for dice objects.

Description

Prints a summary of a dice object.

Usage

## S3 method for class 'dice'
print(x, ...)

Arguments

x

Object of class dice.

...

Ignored for now.

Print method for ice objects.

Description

Prints a summary of an ice object.

Usage

## S3 method for class 'ice'
print(x, ...)

Arguments

x

Object of class ice.

...

Ignored for now.

Summary function for dice objects.

Description

Alias of print method.

Usage

## S3 method for class 'dice'
summary(object, ...)

Arguments

object

Object of class dice.

...

Ignored for now.

Summary function for ice objects.

Description

Alias of print method.

Usage

## S3 method for class 'ice'
summary(object, ...)

Arguments

object

Object of class ice.

...

Ignored for now.

Data concerning white wine.

Description

The WhiteWine data frame has 4898 rows and 12 columns and concerns white wines from a region in Portugal. The response variable, quality, is a wine quality metric, taken to be the median preference score of three blind tasters on a scale of 1-10. The 11 covariates are physicochemical metrics of wine quality such as citric acid content, sulphates, etc.

Usage

data(WhiteWine)

Format

A data frame of 4898 cases on 12 variables.

Source

K Bache and M Lichman. UCI machine learning repository, 2013. http://archive.ics.uci.edu/ml