skbel.goggles

This package contains the skbel.goggles module, which provides some useful visualizations for the skbel package.

skbel.goggles.visualization

Some visualization utilities.

skbel.goggles.visualization._cca_plot(X_scores, Y_scores, X_obs: array = None, Y_obs: array = None, samples=None, sdir: str = None, show: bool = False, annotation_callback=None, mode=None)

Plot the Canonical Variate Pairs.

Parameters:

• mode – mode of inference

• X_scores – Canonical variates for X

• Y_scores – Canonical variates for Y

• X_obs – The X observations.

• Y_obs – The Y observations.

• samples – The samples to plot.

• sdir – The directory to save the plot to.

• show – Whether to show the plot.

• annotation_callback – A callback function to annotate the plot.

skbel.goggles.visualization._despine(fig=None, ax=None, top=True, right=True, left=False, bottom=False, offset=None, trim=False)

Remove the top and right spines from plot(s).

Parameters:

• fig – Figure to despine all axes of, defaults to the current figure.

• ax – Specific axes object to despine. Ignored if fig is provided.

• bottom (top, right, left,) – If True, remove that spine.

• offset – Absolute distance, in points, spines should be moved away from the axes (negative values move spines inward). A single value applies to all spines; a dict can be used to set offset values per side.

• trim – If True, limit spines to the smallest and largest major tick on each non-despined axis.

skbel.goggles.visualization._my_alphabet(az: int)

Method used to make custom figure annotations.

Parameters:

az – Index of the alphabet

Returns:

corresponding letter

skbel.goggles.visualization._proxy_annotate(annotation: list = None, loc: int = 1, fz: float = 11, obj=None)

Places annotation (or title) within the figure box.

Parameters:

• annotation – Must be a list of labels even of it only contains one label. Savvy ?

• fz – Font size

• loc – Location (default: 1 = upper right corner, 2 = upper left corner)

skbel.goggles.visualization._proxy_legend(legend1: legend = None, colors: list = None, labels: list = None, loc: int = 4, marker: list = None, pec: list = None, fz: float = 11, fig_file: str = None, extra: list = None, obj=None)

Add a second legend to a figure @ bottom right (loc=4) https://stackoverflow.com/questions/12761806/matplotlib-2-different-legends-on-same-graph

Parameters:

• legend1 – First legend instance from the figure

• colors – List of colors

• labels – List of labels

• loc – Position of the legend

• marker – Points ‘o’ or line ‘-’

• pec – List of point edge color, e.g. [None, ‘k’]

• fz – Fontsize

• fig_file – Path to figure file

• extra – List of extra elements to be added on the final figure

skbel.goggles.visualization._yield_alphabet(start=0)

Yields the alphabet from a given index.

Parameters:

start – Index of the first letter

skbel.goggles.visualization.cca_vision(X_scores=None, Y_scores=None, X_obs: array = None, Y_obs: array = None, samples=None, n_cut=None, cplot=True, annotation_call=None, fig_dir: str = None, show: bool = False)

Visualize the CCA results.

Parameters:

• X_scores – CCA scores for X

• Y_scores – CCA scores for Y

• X_obs – X observations

• Y_obs – Y observations

• samples – Samples from the model

• n_cut – Only show the first n_cut components

• cplot – Plot the CCA correlation coefficients

• annotation_call – Annotation callback

• fig_dir – Base directory path

• show – Show figure

skbel.goggles.visualization.explained_variance(n_components, evr, n_cut: int = 0, annotation: list = None, fig_file: str = None, show: bool = False, **kwargs)

PCA explained variance plot.

Parameters:

• n_components – Number of components

• evr – Explained variance ratio

• n_cut – Number of components to display

• annotation – List of annotation(s)

• fig_file – Path to figure file

• show – Show figure

skbel.goggles.visualization.pca_scores(training: array, prediction: array = None, pc_post: array = None, random_pcs: array = None, n_comp: int = 0, annotation: list = None, title: str = None, xlabel: str = None, ylabel: str = None, fig_file: str = None, add_legend: bool = True, show: bool = False)

PCA scores plot, displays scores of observations above those of training.

Parameters:

• pc_post – PCA scores of the posterior

• training – Training scores

• prediction – Test scores

• pc_post – PCA scores of the posterior (Y)

• random_pcs – Random PCA scores

• n_comp – How many components to show

• annotation – List of annotation(s)

• title – Title of the plot

• xlabel – Label of the x axis

• ylabel – Label of the y axis

• fig_file – Path to figure file

• add_legend – Add legend

• show – Show figure