plotHistogram

Visualize histogram for a variable in drift detection

Syntax

plotHistogram(DDiagnostics)

plotHistogram(DDiagnostics,Variable=variable)

plotHistogram(ax,___)

H = plotHistogram(___)

Description

plotHistogram(DDiagnostics) plots a histogram of the baseline and target data for the variable with the lowest p-value.

If you set the value of 'EstimatePValues' to false in the call to detectdrift, then plotHistogram displays NaN for the p-value and the drift status.

plotHistogram(DDiagnostics,Variable=variable) plots the histogram of the baseline and target data for the variable specified by variable.

plotHistogram(ax,___) plots into axes ax instead of gca.

H = plotHistogram(___) returns an array of Histogram objects in H. Use this to inspect and adjust the properties of the histogram. For more information on the Histogram object properties, see Histogram Properties.

Examples

Plot Histogram for Variable with Lowest p-Value

Generate baseline and target data with three variables, where the distribution parameters of the second and third variables change for target data.

rng('default') % For reproducibility
baseline = [normrnd(0,1,100,1),wblrnd(1.1,1,100,1),betarnd(1,2,100,1)];
target = [normrnd(0,1,100,1),wblrnd(1.2,2,100,1),betarnd(1.7,2.8,100,1)];

Perform permutation testing for all variables to check for any drift between the baseline and target data.

DDiagnostics = detectdrift(baseline,target)

DDiagnostics = 
  DriftDiagnostics

              VariableNames: ["x1"    "x2"    "x3"]
       CategoricalVariables: []
                DriftStatus: ["Stable"    "Drift"    "Warning"]
                    PValues: [0.3850 0.0050 0.0910]
        ConfidenceIntervals: [2x3 double]
    MultipleTestDriftStatus: "Drift"
             DriftThreshold: 0.0500
           WarningThreshold: 0.1000


  Properties, Methods

Plot the histogram for the default variable.

plotHistogram(DDiagnostics)

plotHistogram by default plots the histogram of the baseline and target data for the variable with the lowest p-value. It also displays the p-value and the drift status for that variable.

Plot Histogram of All Variables in Tiled Layout

Generate baseline and target data with three variables, where the distribution parameters of the second and third variables change for target data.

rng('default') % For reproducibility
baseline = [normrnd(0,1,100,1),wblrnd(1.1,1,100,1),betarnd(1,2,100,1)];
target = [normrnd(0,1,100,1),wblrnd(1.2,2,100,1),betarnd(1.7,2.8,100,1)];

Perform permutation testing for all variables to check for any drift between the baseline and target data. Use the Energy statistic as the metric.

DDiagnostics = detectdrift(baseline,target,"ContinuousMetric","energy")

DDiagnostics = 
  DriftDiagnostics

              VariableNames: ["x1"    "x2"    "x3"]
       CategoricalVariables: []
                DriftStatus: ["Stable"    "Drift"    "Warning"]
                    PValues: [0.3790 0.0110 0.0820]
        ConfidenceIntervals: [2x3 double]
    MultipleTestDriftStatus: "Drift"
             DriftThreshold: 0.0500
           WarningThreshold: 0.1000


  Properties, Methods

Plot the histograms for all three variables in a tiled layout.

tiledlayout(3,1);
ax1 = nexttile;
plotHistogram(DDiagnostics,ax1,Variable="x1")
ax2 = nexttile;
plotHistogram(DDiagnostics,ax2,Variable="x2")
ax3 = nexttile;
plotHistogram(DDiagnostics,ax3,Variable="x3")

Plot Histogram for Drift Detection and Change Bar Color

Generate baseline and target data with three variables, where the distribution parameters of the second and third variables change for target data.

rng('default') % For reproducibility
baseline = [normrnd(0,1,100,1),wblrnd(1.1,1,100,1),betarnd(1,2,100,1)];
target = [normrnd(0,1,100,1),wblrnd(1.2,2,100,1),betarnd(1.7,2.8,100,1)];

Perform permutation testing for all variables to check for any drift between the baseline and target data.

DDiagnostics = detectdrift(baseline,target)

DDiagnostics = 
  DriftDiagnostics

              VariableNames: ["x1"    "x2"    "x3"]
       CategoricalVariables: []
                DriftStatus: ["Stable"    "Drift"    "Warning"]
                    PValues: [0.3850 0.0050 0.0910]
        ConfidenceIntervals: [2x3 double]
    MultipleTestDriftStatus: "Drift"
             DriftThreshold: 0.0500
           WarningThreshold: 0.1000


  Properties, Methods

Plot histogram for the first variable.

H = plotHistogram(DDiagnostics,Variable=1)

H = 
  2x1 Bar array:

  Bar    (Baseline)
  Bar    (Target)

You can access H in the workspace by double clicking on it and change Bar object properties. You can also make changes programmatically. For example, change the color of the histogram bars for baseline data.

H(1).FaceColor = [1 0 1];

Input Arguments

DDiagnostics — Diagnostics of the permutation testing for drift detection

DriftDiagnostics object

Diagnostics of the permutation testing for drift detection, specified as a DriftDiagnostics object returned by detectdrift.

variable — Variable for which to plot histogram

string | character vector | integer index

Variable for which to plot the histogram, specified as a string, a character vector, or an integer index.

Example: Variable="x2"

Example: Variable=2

Data Types: single | double | char | string

ax — Axes to plot into

Axes object | UIAxes object

Axes for plotHistogram to plot into, specified as an Axes or UIAxes object. If you do not specify ax, then plotHistogram creates the plot using the current axes. For more information on creating an axes object, see axes and uiaxes.

Algorithms

• For categorical data, detectdrift adds 0.5 correction factor to histogram bin counts for each bin to handle empty bins (categories). This is equivalent to the assumption that the parameter p, probability that value of the variable would be in that category, has the prior distribution Beta(0.5,0.5), i.e. Jefferys prior assumption for the distribution parameter.

• plotHistogram treats a variable as ordinal for visualization purposes under any of the following cases:

  • If the variable is ordinal in either baseline or target data and the categories from baseline and target data are the same.

  • If the variable is ordinal in either in baseline or target data and the categories of the other data set is a subset of the ordinal data.

  • If the variable is ordinal in both baseline and target data and categories from either one is a subset of the other.

• If a variable is ordinal, plotHistogram preserves the order of the bin names.

Version History

Introduced in R2022a