Skip to content

Commit

Permalink
Improve computation time of compute_matrix for parametric ops (#2759)
Browse files Browse the repository at this point in the history
* separated out speedup PR

* revert IsingXY change

* alternative version

* black

* more working stuff?

* tmp

* converged RZ.compute_matrix

* converged compute_eigvals

* tests

* converged matrices and eigvals for diagonal operations

* compatibility

* test coverage

* isingxy registration

* test attrs

* isingxy: additional tests and compatibility of compute_eigvals

* fix U1 matrix
  • Loading branch information
dwierichs authored Jul 27, 2022
1 parent c8bf6a2 commit 465642b
Show file tree
Hide file tree
Showing 5 changed files with 410 additions and 108 deletions.
4 changes: 4 additions & 0 deletions doc/releases/changelog-dev.md
Original file line number Diff line number Diff line change
Expand Up @@ -268,6 +268,10 @@ of operators. [(#2622)](https://github.com/PennyLaneAI/pennylane/pull/2622)

<h3>Improvements</h3>

* A small performance upgrade to the `compute_matrix` method
of broadcastable parametric operations.
[(#2726)](https://github.com/PennyLaneAI/pennylane/pull/2726)

* Jacobians are cached with the Autograd interface when using the
parameter-shift rule.
[(#2645)](https://github.com/PennyLaneAI/pennylane/pull/2645)
Expand Down
1 change: 1 addition & 0 deletions pennylane/ops/qubit/attributes.py
Original file line number Diff line number Diff line change
Expand Up @@ -223,6 +223,7 @@ def __contains__(self, obj):
"IsingXX",
"IsingYY",
"IsingZZ",
"IsingXY",
]
)
"""Attribute: Operations that support parameter broadcasting.
Expand Down
Loading

0 comments on commit 465642b

Please sign in to comment.