|---| | [ View source on GitHub](https://github.com/tensorflow/probability/blob/v0.25.0/tensorflow_probability/python/experimental/distribute/__init__.py) | Experimental module for doing distributed log prob calculations. ## Classes [`class JointDistributionCoroutine`](https://www.tensorflow.org/probability/api_docs/python/tfp/experimental/distribute/JointDistributionCoroutine): A sharding-aware JointDistributionCoroutine. [`class JointDistributionNamed`](https://www.tensorflow.org/probability/api_docs/python/tfp/experimental/distribute/JointDistributionNamed): A sharding-aware JointDistributionNamed. [`class JointDistributionSequential`](https://www.tensorflow.org/probability/api_docs/python/tfp/experimental/distribute/JointDistributionSequential): A sharding-aware JointDistributionSequential. [`class Sharded`](https://www.tensorflow.org/probability/api_docs/python/tfp/experimental/distribute/Sharded): A meta-distribution meant for use in an SPMD distributed context. ## Functions [`make_pbroadcast_function(...)`](https://www.tensorflow.org/probability/api_docs/python/tfp/experimental/distribute/make_pbroadcast_function): Constructs a function that broadcasts inputs over named axes. [`make_psum_function(...)`](https://www.tensorflow.org/probability/api_docs/python/tfp/experimental/distribute/make_psum_function): Constructs a function that broadcasts inputs over named axes. [`make_sharded_log_prob_parts(...)`](https://www.tensorflow.org/probability/api_docs/python/tfp/experimental/distribute/make_sharded_log_prob_parts): Constructs a log prob parts function that all-reduces over terms.