numpyro
numpyro copied to clipboard
pyro-ppl
Probabilistic programming with NumPy powered by JAX for autograd and JIT compilation to GPU/TPU/CPU.
Currently, Mixed HMC only supports sequential and parallel chain_methods, otherwise an error is raised (#1073). To make Mixed HMC utilize GPUs more efficiently I would like to see the addition...
Resolves https://github.com/pyro-ppl/numpyro/issues/417. This PR tracks the progress of using flow hmc in covtype dataset. ### Problem setting - Data is randomly split into train set with 400,000 data points and...
The recently proposed [NEP-47](https://numpy.org/neps/nep-0047-array-api-standard.html) attempts to unify the APIs of various tensor frameworks (NumPy, Tensorflow, PyTorch, Dask, JAX, CuPy, MXNet, etc.), via the [Python array API standard](https://data-apis.github.io/array-api/latest). It is a...
Hi! This PR adds init_to_mean function - also not sure where to add tests. What do you think?
Currently, we have `scan` implemented in [contrib.control_flow](https://github.com/pyro-ppl/numpyro/blob/master/numpyro/contrib/control_flow/scan.py). After #878, it should be doable to use other control flows in HMC/NUTS. The implementation should mimic what we currently have in `scan`...
Hi, I have started implementing an Empirical distribution by porting the pyro one. Kind of unsure how to do the log probability, I actually think the implementation in pyro might...
It would be great to add BART to the list of models available via numpyro: https://github.com/pymc-devs/pymc3/pull/4183 Happy to take a stab at it given some time.
Hi! I would like to request some guidelines on how to implement SparseGP (with SVI) in Numpyro as already included in [Pyro](https://pyro.ai/examples/gp.html#Sparse-GPs). Currently there is only a simple Gaussian Process...
I was wondering whether there's any way of automatically rendering defined models, similar to [`pymc3.model_graph.model_to_graphviz`](https://docs.pymc.io/api/model_graph.html). If there's currently no way of doing this, I'd be happy to submit a PR...
I noticed that samples from the `AutoDelta` guide don't contain deterministic transforms of random variables in the model. Here's a minimal example which reproduces the issue: ```Python import numpyro import...
