Add SparseObservable.compose
#13766
Draft
Add SparseObservable.compose
#13766
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This adds matrix multiplication between `SparseObservable`s (or one `SparseObservable` and anything that can be coerced to an observable), using the same signature as similar methods in `quantum_info`. The current implementation attempts to be relatively memory efficient, though there are specialisations that may be implemented if we wanted to be faster: - if all the contained terms are Paulis, so each pairwise multiplication produces either zero or one alphabet letters, we currently have a redundant copy-in and copy-out of the Cartesian-product generator that effectively doubles the overhead. We could check whether each term contains only Paulis once, and then avoid that overhead in those cases. - alternatively, the Cartesian-product iterator could be taught to use the buffers of the output `SparseObservable` directly as its output space, rather than requiring copy-out of its buffers, which would remove the overhead from all calls, not just the single-Pauli ones, but requires a bit more bookkeeping, since the references would keep need to be updated. - the case of `qargs` being not `None` and `front=True` currently involves an extra copy step to simplify the logic. If this is an important case, we could add a right-matmul specialisation of `compose_map`. This commit currently does not contain tests, because we currently don't have suitable methods to test the validity of the output without coupling to the particular choice of factorisation of the matrix multiplication. We need either a way to convert to a Pauli-only representation (with the exponential explosion that entails) or to a matrix; either of these can be uniquely canonicalised.
jakelishman
added
Changelog: New Feature
Before Rust 1.83, `const` functions can't borrow `static`s, even immutably and without interior mutability.o There's no need for `matmul` to be `const`; it's always inlined anyway, and `matmul_generate` can be used in `const` contexts.
Pull Request Test Coverage Report for Build 13058536869Details
💛 - Coveralls |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
This adds matrix multiplication between
SparseObservables (or oneSparseObservableand anything that can be coerced to an observable), using the same signature as similar methods inquantum_info.The current implementation attempts to be relatively memory efficient, though there are specialisations that may be implemented if we wanted to be faster:
if all the contained terms are Paulis, so each pairwise multiplication produces either zero or one alphabet letters, we currently have a redundant copy-in and copy-out of the Cartesian-product generator that effectively doubles the overhead. We could check whether each term contains only Paulis once, and then avoid that overhead in those cases.
alternatively, the Cartesian-product iterator could be taught to use the buffers of the output
SparseObservabledirectly as its output space, rather than requiring copy-out of its buffers, which would remove the overhead from all calls, not just the single-Pauli ones, but requires a bit more bookkeeping, since the references would keep need to be updated.the case of
qargsbeing notNoneandfront=Truecurrently involves an extra copy step to simplify the logic. If this is an important case, we could add a right-matmul specialisation ofcompose_map.Details and comments
This commit currently does not contain tests, because we currently don't have suitable methods to test the validity of the output without coupling to the particular choice of factorisation of the matrix multiplication. We need either a way to convert to a Pauli-only representation (with the exponential explosion that entails) or to a matrix; either of these can be uniquely canonicalised.
WIP until it has tests.
Close #13390