Cross-cuts of conditional distributions in 3-d

Conditioning is a way to reduce the dimensionnality of a multivariate distribution. It allows to plot the d

import openturns as ot
import otbenchmark as otb
import matplotlib.pyplot as plt

distribution = ot.Normal(3)

referencePoint = distribution.getMean()

Print the iso-values of the distribution

inputDimension = distribution.getDimension()
description = distribution.getDescription()

crossCut = otb.CrossCutDistribution(distribution)
fig = crossCut.drawConditionalPDF(referencePoint)
# Remove the legend labels because there
# are too many for such a small figure
for ax in fig.axes:
    ax.legend("")
# Increase space between sub-figures so that
# there are no overlap
plt.subplots_adjust(hspace=0.4, wspace=0.4)

Descr =  1 0
Descr =  2 0
Descr =  2 1

fig = crossCut.drawMarginalPDF()
# Remove the legend labels because there
# are too many for such a small figure
for ax in fig.axes:
    ax.legend("")
# Increase space between sub-figures so that
# there are no overlap
plt.subplots_adjust(hspace=0.4, wspace=0.4)