New Squin statements. #200
Conversation
Codecov ReportAttention: Patch coverage is
📢 Thoughts on this report? Let us know! |
☂️ Python Coverage
Overall Coverage
New FilesNo new covered files... Modified Files
|
|
@weinbe58 is there a reason we don't want broadcast in the wire dialect as well? Or is the idea that there will be some lowering from a qubit broadcast to a bunch of wire applications? |
| @@ -138,6 +146,18 @@ class PauliOp(ConstantUnitary): | |||
| pass | |||
|
|
|||
|
|
|||
| @statement(dialect=dialect) | |||
| class CliffordString(ConstantUnitary): | |||
There was a problem hiding this comment.
What is a CliffordString supposed to represent? PauliString has specific uses like defining a multi-qubit operator for measurement, control, or phase rotation. I am not familiar with "Clifford string".
There was a problem hiding this comment.
I wanted to represent a product of Clifford operators not just Pauli. But if this is confusing we don't need to have it.
There was a problem hiding this comment.
Clifford include also CX, do we also want product of mixing 2Q and 1Q?
Some initial friction points/anticipated changes I can immediately see in translating this to stim - The original `PauliChannel` definition from @kaihsin can take a variable number of float arguments, but in the stim dialect there are explicit fields for each probability of a 1Q/2Q pauli operator being applied. My feeling is that this should be reflected on the squin side as just something like `1QPauliChannel` and `2QPauliChannel` with explicit fields. I don't think we'd lose any generality here along with the added benefit that it would be much harder for a user to misinput information (the default definition assumes you plug in either 3 or 15 values via varargs but it seems to easy to make a mistake here) - ~~If we DO want varargs I'm not quite sure how to nicely feed this to the kirin `@wraps` decorator~~ - From #200 , I see it could be possible to not have to provide an operator for `basis` and instead use a `CliffordString` (although I'd have to restrict CliffordString even further, to just the Pauli Operators. Would it be worth having something like a `PauliString`? Or is that silly?) cc: @weinbe58 @Roger-luo @kaihsin --------- Co-authored-by: Phillip Weinberg <[email protected]>
This PR is addressing the new squin statements. However, after discussion with @kaihsin and @Roger-luo we have slightly modified the semantics to make things a bit cleaner:
Boradcastis now an appliy-like statement, see Broadcast should be an Apply-like statement. #206PualiStringis nowCliffordStringto enable better rewrites to STIM.