-
Implements new ansatzs and new methods to compute
-
We probably want to allow advanced users to date using variational_gamma with tree seqs with unary nodes, but we want to fail or warn the casual user that this could be bad. What should the API for th…
-
# 🐛 Bug
I'd like to use natural gradient descent together with the orthogonally decoupled variational strategy, so I ran the [NGD example notebook](https://github.com/cornellius-gp/gpytorch/blob/ma…
-
> [!NOTE]
> If you have a request to support a specific method, or would like to see priority of one of the listed methods, please open a separate issue, so it won't get buried in this thread. Base…
-
Once we have a reasonable variational method working, we should add a test or two to `test_accuracy.py` so that we pick up any regressions. Do you think you could do this @nspope ?
-
As someone involved in the recent tsdate releases, I'm wondering how tsinfer+tsdate_v0.2.1 performs (e.g. compared to Relate). The recommended way to perform ARG inference now is
```python
import …
-
It might be useful to add a dedicated `approximate(Ψ₀::AbstractMPS, Ψ::AbstractMPS, algorithm, [environments]; kwargs...)` method for a globally optimal truncation of a given MPS. It is mentioned in t…
-
### Describe the workflow you want to enable
Currently, there is no included implementation of a PCA algorithm made for handling binary data in the scikit-learn library. However, the algorithm for …
-
When I run the code in the "complete training loop",
it has the following errorr message:
---------------------------------------------------------------------------
AttributeError …
-
The API description has the following two methods for `vi`:
- `vi(model, alg)`
- `vi(model, alg, q::VariationalPosterior, θ)`
The first method is needed when the algorithm only supports one varia…