Skip to content

Tags: aesara-devs/aeppl

Tags

v0.1.5

Toggle v0.1.5's commit message
Add a deprecation warning filter for newer versions of NumPy

Earlier versions of Aesara still use the deprecated `np.product` function.

v0.1.4

Toggle v0.1.4's commit message
Preserve the identities of valued/observed variables

v0.1.3

Toggle v0.1.3's commit message
Fix ValueVariable checks

v0.1.2

Toggle v0.1.2's commit message
Explicitly compile functions in scipy_logprob_tester and update test_…

…vonmises_logprob

v0.1.1

Toggle v0.1.1's commit message
Explicitly remove AePPL IR from conditional_logprob results

v0.1.0

Toggle v0.1.0's commit message
Convert xfail test into a standard unit test

v0.0.50

Toggle v0.0.50's commit message
Refactor MeasurableElemwise and automatic random variable transforms

These changes make it possible for users to extend coverage of measurable
`Elemwise` transforms.  They also put these rewrites under the management of
Aesara's rewrite system, which means they can be applied more efficiently and
their use can customized and tracked with more granularity.

`sub`, `neg`, and `true_div` support is added, as well.

v0.0.40

Toggle v0.0.40's commit message
Associate transforms with random variables

v0.0.39

Toggle v0.0.39's commit message
Implement log-probability for StudentTRV

v0.0.38

Toggle v0.0.38's commit message
Fix remaining ShapeFeature import