Matthew Treinish

I ran asv on this and none of the benchmarks showed an improvement or regression beyond system noise. I did see up to a 10% improvement doing local testing with...

I think if they're general synthesis (where they go from a `quantum_info` object or a matrix) to a circuit they should live in `qiskit.quantum_info.synthesis`. Ideally I feel like the `qiskit.extensions`...

One thought I did have is that we could do some of the work in parallel if we allow the synthesis function to take a list of unitaries as an...

I tend to agree, the 2q distance does seem like overkill for determining that a trivial layout is perfect. `CheckMap` is more than enough to determine if the trivial layout...

> Probably it's not a problem, since numpy in Rust can have `PyObject`s as elements. The bigger issue with this is that to do any numeric operations on an ndarray...

FWIW, I'm not sure how much this is actually needed. The cx count and measurement count are only used as weighting factors when trying to estimate how noisy the qubits...

I separated out the fix for the initial layout permutation in #8802 we should probably rebase this on top of that since it changes the first half of the code...

I pivoted the PR to have the hook point methods to return a method name string to be used as the default method when calling transpile on the backend in:...

I've marked this as on hold because I'd like to wait until #8051 merges so we have a rust bin->hex converter so we can add an option to the rust...

I've removed the on hold from this, I was misremembering #8051 and it doesn't have an efficient rust bin to hex converter (it has a hex to bin converter). Let...