Graham

It's not as well documented as it used to be, but [the `pyquil.latex`](https://pyquil-docs.rigetti.com/en/v2.28.2/apidocs/visualization.html) section could [be useful](https://github.com/rigetti/pyquil/tree/master/pyquil/latex). Note the warning: > For this to work, you need two external programs,...

Sorry, @BoltzmannEntropy! I was confused by the circuit diagram in the “or” section of your note.

See #1465

Some benchmarks might want to look at readout fidelity or some such for qubits that don’t happen to have any gates available, I suppose.

Here's a workaround/suggestion for what the proposed optional argument might instigate: ```python import networkx as nx from pyquil import get_qc computer = get_qc() qubits = computer.qubits() topology = computer.qubit_topology() isa...

Code golfed version which may work for some scenarios: ```python computer = get_qc() qubits = [q.id for q in computer.to_compiler_isa().qubits.values() if not q.dead] topology = computer.qubit_topology().subgraph(live_qubits) ```

Seconded! I'd really like this feature—though I understand that this is a (wonderful!) open source library, and I'm lucky that it exists at all, and I shouldn't be demanding features.

`qutip` has a [`simdiag` function](https://qutip.org/docs/latest/apidoc/functions.html#qutip.simdiag), though [others have noted some numerical instability](https://github.com/qutip/qutip/issues/756).

There's also [NUMERICAL METHODS FOR SIMULTANEOUS DIAGONALIZATION](https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.55.1799&rep=rep1&type=pdf), which [this rando Matlab](https://www.mathworks.com/matlabcentral/fileexchange/46794-simdiag-m) file claims to implement.

Since `U U^T` is symmetric, Golub and Van Loan (pg. 500 in the 4th edition) may have what we need (though I don't know if `B = Im{U U^T}` would...