rc.gsa.base

Base Classes and basic functions underpinning GSA.

Classes

Calibrator	Interface to GSA calibrator
Gaussian	Encapsulates a Gaussian pdf. For numerical stability the 2 Pi factor is not included.

Methods

diag_det(tensor)

Determinant of a diagonal tensor.

Module Contents

diag_det(tensor)

Determinant of a diagonal tensor.

Parameters:

tensor (TF.Tensor) – Of shape [...,m]. The last axis must contain the diagonal

Returns: Tensor shaped [...].

class Calibrator

Bases: abc.ABC

Interface to GSA calibrator

class Gaussian(mean, variance, is_variance_diagonal, ordinate=tf.constant(0, dtype=Float()), LBunch=2)

Encapsulates a Gaussian pdf. For numerical stability the 2 Pi factor is not included.

Parameters:

• mean (TF.Tensor)

• variance (TF.Tensor)

• is_variance_diagonal (bool)

• ordinate (TF.Tensor)

• LBunch (int)

property det: TF.Tensor

The sqrt of the determinant of the Gaussian covariance.

Return type:

TF.Tensor

property pdf: TF.Tensor

Calculate the Gaussian pdf from the output of Gaussian.

Return type:

TF.Tensor

expand_dims(axes)

Insert dimensions at the specified axes.

Parameters:

axes (rc.base.definitions.Sequence[int]) – A sequence of dims to insert.

Return type:

Gaussian

Returns: self for chaining calls.

__truediv__(other)

Divide this Gaussian pdf by denominator.

Parameters:

other – The Gaussian to divide by.

Return type:

Gaussian