viziphant.gpfa.plot_cumulative_shared_covariance

viziphant.gpfa.plot_cumulative_shared_covariance(loading_matrix)[source]

This function plots the cumulative shared covariance. It allows to visually identify an appropriate number of dimensions which is small on the one hand, but explains a substantial part of the variance in the data on the other hand.

Parameters:
loading_matrixnp.ndarray

The loading matrix defines the mapping between neural space and latent state space. It is obtained by fitting a GPFA model and stored in GPFA.params_estimated['C'] or if orthonormalized GPFA.params_estimated['Corth'].

Returns:
figmatplotlib.figure.Figure
axesmatplotlib.axes.Axes