-
### Describe the issue:
```
ERROR (pytensor.graph.rewriting.basic): Rewrite failure due to: constant_folding
ERROR (pytensor.graph.rewriting.basic): node: Cast{float64}(0)
ERROR (pytensor.graph.re…
-
Not quite sure if this will be fixed in the upcoming `0.7.0` release but noticed the re-written [BudgetOptimizer](https://github.com/pymc-labs/pymc-marketing/blob/0144bd37d692d43f8575ec2c522289b438f22…
-
### Description
Make import `pytensor as pt` basically work like today's `import pytensor.tensor as pt`.
For users that's all they need, besides functionality like `scan`, `ifelse`, `function`,…
-
### Description
Brought up in #846
```python
import pytensor
import pytensor.tensor as pt
x = pt.vector("x", dtype="int64")
out = pt.special.softmax(x)
# Doesn't seem right
out.dprint…
-
### Describe the issue:
I'm performing (offline) Bayesian Optimization using pymc Marginal GP. Maximizing the (expected improvement) acquisition function requires making repeated predictions of the G…
-
When applying the `linlog` model for use in `pymc` inference, we face quite significant inference time (2 days - 10 days). This is obviously not ideal for any productive application of the workflow. U…
-
This new release has breaking changes because of the latest version of PyTensor. We are "solving" this by having an upper bound on the PyMC version, see https://github.com/pymc-labs/pymc-marketing/pul…
-
In our previous meeting, Jeremy mentioned that emll now automatically handles missing data.
I'm using emll right now, and it's complaining about array shape mismatches, but I have no idea where thes…
-
https://github.com/pymc-devs/pytensor/blob/5d4b0c4b9a1e478dda48e912ee708a9e557e9343/pytensor/tensor/random/basic.py#L897-L903
It is not allowed to broadcast with mu so it should be aware of its sha…
-
The barebones PyTorch backend should be in in the next PyTensor release and subsequent PyMC.
I think the best way to test it out and gather info on what Ops are most needed is to give access from n…