dreye.compute_mean_width#

dreye.compute_mean_width(X, n=1000, vectorized=False, center=False, seed=None)[source]#

Compute mean width of a matrix X by projecting it onto random vectors.

Parameters:
Xnumpy.ndarray

An n x m matrix with n samples and m features.

nint, optional

The number of random projections to calculate width, by default 1000.

vectorizedbool, optional

If True, use vectorized operations for speed up, by default False.

centerbool, optional

If True, center the data before computing, by default False.

seedint, optional

The seed for the random number generator, by default None.

Returns:
float

The mean width of X.