numpyro icon indicating copy to clipboard operation
numpyro copied to clipboard

Probabilistic programming with NumPy powered by JAX for autograd and JIT compilation to GPU/TPU/CPU.

Results 154 numpyro issues
Sort by recently updated
recently updated
newest added

Requires #1892

awaiting response

This PR refactors the build infrastructure to 1. remove the setup.py in favor of pyproject.toml 2. dynamically generate the version from the pyproject (this is not a VCS dynamic versioning)...

The `MaskedDistribution` object wraps models, but does not necessarily expose their public API. I'm raising this Issue to perhaps add support for the `components_` methods. Possible Options: 1. Some clever...

question

I'm assuming these functions are expected to work with seeded models, but they yield ``UnexpectedTracerError``s. ```python import numpyro import jax import jax.random as jr import numpyro.distributions as dist from numpyro.infer.inspect...

documentation

Jax-0.4.31: Runtime: 27.06 seconds https://colab.research.google.com/drive/1EsFY1St8Y2ZNBZ9UXTa9FDWrjPDdTU4U?usp=sharing Jax-0.4.33: Runtime: 84.91 seconds https://colab.research.google.com/drive/1g7GkuK4-GloO6cywvDUf5BVU9qO2jf1W?usp=sharing I’m not sure if this issue is specific to ```flax_random_module``` or a broader problem, but I’ve primarily been using NumPyro...

performance
jax

Hello! I have a custom multi-dimensional distribution where the support may be truncated along some dimensions. In terms of constraints, some dimensions will either be `real`, `greater_than`, `less_than`, or `interval`....

enhancement
good first issue

I have defined some distributions as child classes of `Distribution`. They are very closely linked, and one is used in another definition. I made a mistake not testing them. Upon...

question

See the forum issue https://forum.pyro.ai/t/enumerate-support-for-batch-dimensions-of-custom-distribution/7656/4 We need to move the logic of exclude_deterministic to the `infer_discrete` branch https://github.com/pyro-ppl/numpyro/blob/3cde93d0f25490b9b90c1c423816c6cfd9ea23ed/numpyro/infer/util.py#L793-L818

bug
good first issue

### Changes made Added support for passing a parameter called `initial_value` to `RecursiveLinearTransform` to facilitate forecasting when using the transform for latent variables. ### Links to related PRs #2037 ###...

### Feature Summary Add a parameter called `initial_value` to `RecursiveLinearTransform`. ### Why is this needed? When using RecursiveLinearTransform for modeling structural time series, I sometimes run into limitations related to...

enhancement