Oryx is a library for probabilistic programming and deep learning built on top
of Jax. The approach is to expose a set of function transformations that compose
and integrate with JAX's existing transformations (e.g. jit, grad, and
vmap).
You can install Oryx via pip:
$ pip install oryxDocumentation is available on the Oryx website.
To develop and modify Oryx, you need to install
poetry, a tool for Python packaging and
dependency management.
To install the development dependencies of Oryx, you can run
$ poetry installand to enter a virtual environment for testing or debugging, you can run:
$ poetry shellOryx uses Bazel for building and testing. Once Bazel is installed, you can run tests by executing:
$ poetry run bazel test //oryx/...