Transformations#

Transform Instances#

Transform instances are the entities that should be used in the transform parameter to a random variable constructor.

simplex

Instantiation of pymc.logprob.transforms.SimplexTransform for use in the transform argument of a random variable.

logodds

Instantiation of pymc.logprob.transforms.LogOddsTransform for use in the transform argument of a random variable.

log_exp_m1

Instantiation of pymc.distributions.transforms.LogExpM1 for use in the transform argument of a random variable.

ordered

Instantiation of pymc.distributions.transforms.Ordered for use in the transform argument of a random variable.

log

Instantiation of pymc.logprob.transforms.LogTransform for use in the transform argument of a random variable.

sum_to_1

Instantiation of pymc.distributions.transforms.SumTo1 for use in the transform argument of a random variable.

circular

Instantiation of pymc.logprob.transforms.CircularTransform for use in the transform argument of a random variable.

Specific Transform Classes#

CholeskyCovPacked(n)

Transforms the diagonal elements of the LKJCholeskyCov distribution to be on the log scale

Interval([lower, upper, bounds_fn])

Wrapper around pymc.logprob.transforms.IntervalTransform for use in the transform argument of a random variable.

LogExpM1()

Ordered([ndim_supp])

SumTo1([ndim_supp])

Transforms K - 1 dimensional simplex space (k values in [0,1] and that sum to 1) to a K - 1 vector of values in [0,1] This Transformation operates on the last dimension of the input tensor.

ZeroSumTransform(zerosum_axes)

Constrains any random samples to sum to zero along the user-provided zerosum_axes.

Transform Composition Classes#

Chain(transform_list)