Rémi Louf
Rémi Louf
I think I have found a small API inconsistency. Indeed, while `PermutationRV` behaves identically to its numpy equivalent by interpreting `permutation(a)` as `permutation(np.range(a))` when `a` is an `int`: ```python import...
Interestingly, `ChoiceRV`'s behavior also differs from NumPy's in that it does not accept input tensors that have more than one dimention while NumPy does. The following raises an error: ```python...
Thank you for the thorough review. I included all suggestions that were not related to the behavior when `size=None` and rebased the changes. I think we may define "sample" differently?...
Regarding the docstring for the `size` argument. I think that the following: ```python size Sample shape. If the given size is `(m, n, k)`, then `m * n * k`...
Documentation for the binomial is very clear in NumPy, but I've found that it wasn't always the case. Imo it's ok to document the simplest case in these docstrings, and...
I have updated all the docstrings to only explain the non-broadcasted cases (eg scalar and scalar parameters for the normal distribution, scalar and vector parameters for the multinomial distribution, etc.)...
Let me take a look at this once I'm done with my open PRs. That and `Scan` are in a way of being able to compile Aesara models and samplers...
I’m a beginner gopher, but I would like to avoid adding a layer of python to he existing Go backend just to have sparse matrices. If this is still something...
Out of scope for v1.
Better API documentation: https://bylr.info/articles/2022/05/10/api-doc-with-sphinx-autoapi/