Skip to content

Commit ce15075

Browse files
committed
Improve changelog entry
1 parent b47728a commit ce15075

1 file changed

Lines changed: 7 additions & 6 deletions

File tree

CHANGELOG.md

Lines changed: 7 additions & 6 deletions
Original file line numberDiff line numberDiff line change
@@ -10,12 +10,13 @@ changelog does not include internal changes that do not affect the user.
1010

1111
### Changed
1212

13-
- **BREAKING**: Removed the concepts of generalized Gramian, `PSDTensor`, `GeneralizedWeighting`,
14-
and `Flattening`. The `Engine.compute_gramian` method now always returns a square matrix of shape
15-
`[m, m]`, where `m` is the total number of elements of the ``output`` tensor (treating all
16-
dimensions uniformly). Previously, an output of shape `[m_1, m_2]` would return a 4D generalized
17-
Gramian of shape `[m_1, m_2, m_2, m_1]`; it now returns a `[m_1 * m_2, m_1 * m_2]` matrix. To
18-
update, replace `Flattening(weighting)` with a standard `Weighting` and reshape the resulting
13+
- **BREAKING**: Remove entirely the concept of generalized Gramians. The `Engine.compute_gramian`
14+
method now always returns a square matrix of shape `[m, m]`, where `m` is the total number of
15+
elements of the ``output`` tensor (treating all dimensions uniformly). Previously, an output of
16+
shape `[m_1, m_2]` would return a 4D generalized Gramian of shape `[m_1, m_2, m_2, m_1]`; it now
17+
returns a `[m_1 * m_2, m_1 * m_2]` matrix.
18+
This also removes `GeneralizedWeighting` and `Flattening`.
19+
To update, replace `Flattening(weighting)` with a standard `Weighting` and reshape the resulting
1920
weight vector yourself:
2021
```python
2122
# Before

0 commit comments

Comments
 (0)