Plotters¶

Utilities for plotting quantities computed in active subspaces.

active_subspaces.utils.plotters.eigenvalues(e, e_br=None, out_label=None, opts=None)¶

Plot the eigenvalues with bootstrap ranges.

Parameters:	e (ndarray) – k-by-1 matrix that contains the estimated eigenvalues e_br (ndarray, optional) – lower and upper bounds for the estimated eigenvalues. These are typically computed with a bootstrap. (default None) out_label (str, optional) – a label for the quantity of interest (default None) opts (dict, optional) – a dictionary with some plot options (default None)

See also

utils.plotters.subspace_errors(), utils.plotters.eigenvalues()

Notes

This function will plot at most the first four eigevectors in a four-subplot figure. In other words, it only looks at the first four columns of W.

active_subspaces.utils.plotters.plot_opts(savefigs=True, figtype='.eps')¶

A few options for the plots.

Parameters:	savefigs (bool) – save figures into a separate figs director figtype (str) – a file extention for the type of image to save
Returns:	opts – the chosen options. The keys in the dictionary are figtype, savefigs, and font. The font is a dictionary that sets the font properties of the figures.
Return type:	dict

active_subspaces.utils.plotters.subspace_errors(sub_br, out_label=None, opts=None)¶

Plot the estimated subspace errors with bootstrap ranges.

Parameters:	sub_br (ndarray) – (k-1)-by-3 matix that contains the lower bound, mean, and upper bound of the subspace errors for each dimension of subspace. out_label (str, optional) – a label for the quantity of interest (default None) opts (dict, optional) – a dictionary with some plot options (default None)

See also

utils.plotters.eigenvectors(), utils.plotters.eigenvalues()

active_subspaces.utils.plotters.sufficient_summary(y, f, out_label=None, opts=None)¶

Make a summary plot with the given predictors and responses.

Parameters:	y (ndarray) – M-by-1 or M-by-2 matrix that contains the values of the predictors for the summary plot. f (ndarray) – M-by-1 matrix that contains the corresponding responses out_label (str, optional) – a label for the quantity of interest (default None) opts (dict, optional) – a dictionary with some plot options (default None)

Notes

If y.shape[1] is 1, then this function produces only the univariate summary plot. If y.shape[1] is 2, then this function produces both the univariate and the bivariate summary plot, where the latter is a scatter plot with the first column of y on the horizontal axis, the second column of y on the vertical axis, and the color corresponding to f.

active_subspaces.utils.plotters.zonotope_2d_plot(vertices, design=None, y=None, f=None, out_label=None, opts=None)¶

A utility for plotting (m,2) zonotopes with designs and quadrature rules.

Parameters:

vertices (ndarray) – M-by-2 matrix that contains the vertices that define the zonotope
design (ndarray, optional) – N-by-2 matrix that contains a design-of-experiments on the zonotope. The plot will contain the Delaunay triangulation of the points in design and vertices. (default None)
y (ndarray, optional) – K-by-2 matrix that contains points to be plotted inside the zonotope. If y is given, then f must be given, too. (default None)
f (ndarray, optional) – K-by-1 matrix that contains a color value for the associated points in y. This is useful for plotting function values or quadrature rules with the zonotope. If f is given, then y must be given, too. (default None)
out_label (str, optional) – a label for the quantity of interest (default None)
opts (dict, optional) – a dictionary with some plot options (default None)

Notes

This function makes use of the scipy.spatial routines for plotting the zonotopes.