probability

probability copied to clipboard

## Short description The [`Mixture`](https://www.tensorflow.org/probability/api_docs/python/tfp/substrates/jax/distributions/Mixture) distribution does not seem to be compatible with tuple-valued distributions, raising an error: ``` TypeError: Dimension value must be integer or None or have an...

pawel-czyz

instructions to build wheel from source not working

1

Hi. The build instructions to build the wheel from source given in the readme file do not work. ```console bazel build --copt=-O3 --copt=-march=native :pip_pkg mkdir -p dist && ./bazel-bin/pip_pkg dist...

gl-yziquel

Fixing paths for python wheel generation.

1

https://github.com/tensorflow/probability/issues/1816

gl-yziquel

The mailing list is full of spam. Please clean it up.

gl-yziquel

MultivariateNormal* constructor crashes with Numpy 2.0

2

I updated to Numpy 2.0 and found that the `MultivariateNormalDiag` and `MultivariateNormalFullCovariance` constructors crashed because `np.issctype` has been removed. Is Numpy 2.0 supported, or will it be soon? Here is...

slinderman

Simple network with fipout layers does not work with mixed precision

1

Dear tfp devs Firstly, thank you for this magnificent package! I have tried using mixed precision the same way as with base tf(it works on tf). Unfortunately an error occurs:...

szperajacyzolw

Hello, I have been trying to use variational inference to the MLP and CNN (conv2D) layers for my research. Unlike a comprehensive discussion and examples for probabilistic MLP layers, I...

haniehmsv

Here is a simple example of an associative function that `scan_associative` fails to handle because it assumes the associative op broadcasts. The solution is to use `jax.vmap` to distributed elements...

Joshuaalbert

Hi all, there are two small issues with the documentation of experimental.sequential.extended_kalman_filter. 1) The code in the example does not run, giving ` InvalidArgumentError: cannot compute Sub as input #1(zero-based)...

beat-247

Error in HamiltonianMonteCarlo when using gather in log_prob function

2

I am trying to run a Hamiltonian MCMC on a target distribution whose implementation involves a call to tf.gather. The following code: ```python import tensorflow as tf import tensorflow_probability as...

martin-wiebusch-thg