-
Eingdecompositon.
Singular value decomposition
chelosky decomposition.
Application of decompostions
-
From #16209 referencing #12711
> I think MatrixDecompositions should be resurrected (along with most of the changes in #12711) and all decomposition should be given a consistent naming scheme (I wo…
-
As described in #2 handling time-varying matrix size would be a valuable addition to this package. On looking into it the required algorithms seem straight-forward, so it may get done in the coming mo…
-
### 🐛 Describe the bug
When trying to execute diag on matrix input it is currently dispatched to diagonal_copy which is not core IR.
```python
import torch
from torch._dynamo.backends.common impor…
-
When `Eigen::SelfAdjointEigenSolver` (called during KAK decomposition) finds repeated eigenvalues in a matrix, the eigenvectors and their order are not uniquely determined. Rounding errors lead to dif…
-
In this ticket, I add implementations for the following decompositions:
* Decompositions for square matrices over non-archimedean local fields:
1. **Iwasawa**.
2. **Cartan**.
3. **Bruhat-Iwahori**.…
-
The following keyframes raise an uncaught TypeError.
```
[{transform: 'translate(50em)'},
{transform: 'translate(200px) rotate(40deg)'}]
```
This is fine:
```
[{transform: 'translate(50px)'},
{tran…
-
**Is your feature request related to a use case or problem? Please describe.**
`two_qubit_matrix_to_cz_operations` and `three_qubit_matrix_to_operations` are used from within the decompose method of …
-
To be able to reuse matrix allocations, it would be nice to have a possibility to calculate matrix decompositions while reusing their already allocated storage. Each decomposition type would have a ne…
-
Parameters passed to H2O-3 backend:
ignored_columns -> null
impute_missing -> true
seed -> 1
ignore_const_cols -> true
max_runtime_secs -> 0.0
score_each_iteration -> false
model_id -> null
pca_metho…