Jesse Grabowski

Expose all `nutpie` compile backends through `pm.sample`

3

### Description Nutpie currently has two compile modes, numba and JAX, with a 3rd pytorch backend on the way. It would be nice if we could easily access these via...

I followed the guide posted by @fonnesbeck [here](https://gist.github.com/fonnesbeck/a9a0b1624324ee2df30ff32c51e452a8), let me know if this is missing anything!

Several keyword arguments in `nutpie.sample` appear to do nothing

6

~The following are listed in the docstring for `sample` but do not appear to be used anywhere:~ - ~num_try_init~ - ~store_divergences~ - ~store_unconstrained~ - ~store_gradient~ - ~store_mass_matrix~ - ~target_accept~ -...

Allow batch dimension on data in statespace models

2

The most requested feature for the statespace module is to handle multiple time series in a single MCMC run. This will require support for batch dimensions. ~~I think the easiest...

A bit of an oversight; this utility function is hard-coded to look for the posterior group. There's no reason why it shouldn't also work on priors.

Very early draft, do not look TODOs (subject to evolution): - [ ] Marginalization of GP random variables using the MarginalModel API - [ ] New model transformation for creating...

Modular model API

1

Alternative model construction API that uses composable components. Examples to come.

For system with many states, kalman filtering is slow because we have to compute the so-called Kalman gain matrix, $$K = P_\text{prior}Z^T F^{-1}$$, with $$F = ZP_{\text{prior}}Z^T + H$$ From...

### Description Reported [here](https://discourse.pymc.io/t/pytensor-errors-from-using-a-sparse-precision-matrix-for-gaussian-data/16773) on the forum. Reproducible example: ```py import pytensor.tensor as pt from pytensor import sparse W_pt = sparse.csc_from_dense(pt.eye(5)) x = pt.dscalar('x') z = (W_pt * x) gz...

Name check on `pt.tensor` is too strict

2

### Description ```py a0 = pt.tensor(name="a0", dtype=floatX, shape=(m,)) ``` Fails with the following error: ```bash a0 = pt.tensor(name="a0", dtype=floatX, shape=(m,)) /home/jesse/mambaforge/envs/pymc-statespace/lib/python3.11/site-packages/pytensor/tensor/type.py:791: in tensor raise ValueError( E ValueError: The first and...