-
A common problem with the new build tools is that they don't seem to support extension modules. This is often a blocker for projects to migrate to those new tools from setuptools. It would be nice to …
-
The following code runs much slower than it's numpy equivalent:
```c++
#include "xtensor/xnoalias.hpp"
#include "xtensor/xtensor.hpp"
#include "xtensor/xarray.hpp"
#include "xtensor/xrandom.hpp…
-
### Describe the issue:
One case where files were missing: #310
The sdist was not finding the correct version number: #293
I think these examples demonstrate the need for some checks regarding…
-
As soon as pymc 4.0.0 is released we should probably contact the developers about including pypesto in the list of software tools using pymc.
-
### Describe the issue:
Not sure if this belongs here or in the pytensor repo. Putting it here because the minimal example I can come up with uses PyMC. If you make a scan variable, register it wit…
-
### Describe the issue:
When calling `pm.Data` on a matrix (`pd.DataFrame`) where all columns are of type `int` it still returns a tensor of type `float64`. This is not true when passing a `pd.Series…
-
The GitHub icon in our docs links to https://github.com/pymc-devs. Irrespective of whether the docs are for PyMC, PyTensor, PyMC-BART, etc. Is this the intended behavior? I find it annoying and potent…
-
Hi, I have been trying to save, reload and generate new predictions with a model that includes a BARTRV.
I am able to save the trace as a pickle (net_cdf works too), and then instantiate a new mod…
-
### Description
The model fgraph functionality (used by `do` and `observe`) doesn't work with GP variables because those are not represented as pure PyTensor objects and make use of variables from …
-
### Before
```python
index = array(['2017-06-02T00:00:00.000000000', '2017-06-03T00:00:00.000000000',
'2017-06-04T00:00:00.000000000', ...,
'2021-06-28T00:00:00.000000000', '2021-0…