Release notes

This is the 0.24.0 release of TensorFlow Probability. It is tested and stable against TensorFlow 2.16.1 and JAX 0.4.25 .

NOTE: In TensorFlow 2.16+, tf.keras (and tf.initializers, tf.losses, and tf.optimizers) refers to Keras 3. TensorFlow Probability is not compatible with Keras 3 -- instead TFP is continuing to use Keras 2, which is now packaged as tf-keras and tf-keras-nightly and is imported as tf_keras. When using TensorFlow Probability with TensorFlow, you must explicitly install Keras 2 along with TensorFlow (or install tensorflow-probability[tf] or tfp-nightly[tf] to automatically install these dependencies.)

Change notes

• TensorFlow Probability now supports Python 3.12.

  • But note that many parts of tfp.layers and tfp.experimental.nn will raise errors because of a TensorFlow + wrapt bug (see tensorflow/tensorflow#60687 ), which can be worked around by setting the environment variable WRAPT_DISABLE_EXTENSIONS=true.

• Added an experimental implementation of Chopin, Jacob, Papaspiliopoulos, "SMC^2: an efficient algorithm for sequential analysis of state-space models", Journal of the Royal Statistical Society Series B: Statistical Methodology 75.3 (2013). See https://github.com/tensorflow/probability/blob/v0.24.0/tensorflow_probability/python/experimental/mcmc/particle_filter.py#L766 .

• Added tfp.experimental.fastgp, a library for approximately training and evaluating Gaussian Processes in sub-O(n^3) time.
  See https://github.com/tensorflow/probability/tree/r0.24/tensorflow_probability/python/experimental/fastgp .

Huge thanks to all the contributors to this release!

• Alessandro Slamitz
• Christopher Suter
• Colin Carroll
• Emily Fertig
• Gareth Williams
• Jacob Burnim
• Jake VanderPlas
• Matthew Feickert
• Pavel Sountsov
• Richard Levasseur
• Srinivas Vasudevan
• Thomas Colthurst
• Urs Köster