Singular Value Decomposition. Factorizes the matrix `a` into two unitary matrices, ``U`` and ``Vh``, and a 1-dimensional array of singul

Cast the type t to either double or cdouble.

Uses ridge regression with a bootstrapped held-out set to get optimal alpha values for each response. [nchunks] random chunks of length [chunklen]

Logs a status and timing update to [logger] every [countevery] draws from [iterable]. If [total] is given, log messages will include the estimated

Uses a look-up table to force the values in [vec] to be gaussian.

Multiply a full matrix by a diagonal matrix. This function should always be faster than dot. Input: d -- 1D (N,) array (contains the di

Uses ridge regression to find a linear transformation of [stim] that approximates [resp]. The regularization parameter is [alpha]. Parameters

Uses ridge regression to find a linear transformation of [Rstim] that approximates [Rresp], then tests by comparing the transformation of [Pstim]

Uses ridge regression to find a linear transformation of [Rstim] that approximates [Rresp], then tests by comparing the transformation of [Pstim]

Centers the rows of [mat] by subtracting off the mean, but doesn't divide by the SD. Can be undone like zscore.

Gaussianizes each column of [mat].

Creates non-interpolated concatenated delayed versions of [stim] with the given [delays] (in samples). If [circpad], instead of being pa

Un-Z-scores the rows of [mat] by multiplying by unzvals[:,0] (the standard deviations) and then adding unzvals[:,1] (the row means).

Z-scores the rows of [mat] by subtracting off the mean and dividing by the standard deviation. If [return_unzvals] is True, a matrix will be r