elisa.util.bslogu#

Bi-Symmetric log transformation for uniform distribution.

Contributed by @xiesl97 (xiesl97).

log(x, c=0.43429448190325176)[source]#: transformation x -> y c default to 1/log(10)

pow(y, c=0.43429448190325176)[source]#: inverse transformation y -> x c default to 1/log(10)

class BiSymTransform(domain=Real(), c=0.43429448190325176)[source]#

Attributes

sign

Sign of the derivative of the transform if it is bijective.

codomain
inv

Methods

`__call__`(x)	Call self as a function.
`forward_shape`(shape)	Infers the shape of the forward computation, given the input shape.
`inverse_shape`(shape)	Infers the shapes of the inverse computation, given the output shape.

class BiSymLogUniform(low, high, *, c=0.43429448190325176, validate_args=None)[source]#

the input should be transfor before input

for example: import numpy as np import bslogu as bs from numpyro.distributions import LogUniform, Uniform

Uniform( low , high ) LogUniform( np.log(low) , np.log(high) ) BiSymLogUniform( bs.log(low) , bs.log(high) )

Attributes

`batch_shape`	Returns the shape over which the distribution parameters are batched.
`event_dim`
`event_shape`	Returns the shape of a single sample from the distribution without batching.
`mean`	Mean of the distribution.
`variance`	Variance of the distribution.

Methods

`__call__`(args, *kwargs)	Call self as a function.
`cdf`(x)	The cumulative distribution function of this distribution.
`entropy`()	Returns the entropy of the distribution.
`enumerate_support`([expand])	Returns an array with shape len(support) x batch_shape containing all values in the support.
`expand`(batch_shape)	Returns a new `ExpandedDistribution` instance with batch dimensions expanded to batch_shape.
`expand_by`(sample_shape)	Expands a distribution by adding `sample_shape` to the left side of its `batch_shape`.
`get_args`()	Get arguments of the distribution.
`icdf`(q)	The inverse cumulative distribution function of this distribution.
`infer_shapes`(args, *kwargs)	Infers `batch_shape` and `event_shape` given shapes of args to `__init__()`.
`mask`(mask)	Masks a distribution by a boolean or boolean-valued array that is broadcastable to the distributions `Distribution.batch_shape` .
`sample`(key[, sample_shape])	Returns a sample from the distribution having shape given by sample_shape + batch_shape + event_shape.
`sample_with_intermediates`(key[, sample_shape])	Same as `sample` except that any intermediate computations are returned (useful for TransformedDistribution).
`shape`([sample_shape])	The tensor shape of samples from this distribution.
`to_event`([reinterpreted_batch_ndims])	Interpret the rightmost reinterpreted_batch_ndims batch dimensions as dependent event dimensions.
`validate_args`([strict])	Validate the arguments of the distribution.

arg_constraints: dict[str, Any] = {'high': Dependent(), 'low': Dependent()}#

The cumulative distribution function of this distribution.

Parameters:: value – samples from this distribution.
Returns:: output of the cumulative distribution function evaluated at value.