Update Optimize1qGatesDecomposition to be Target aware

[ajavadia]

Summary

Updating the Optimize1qGatesDecomposition pass to take into account Target if available. This means that it chooses the best decomposition adaptively per qubit, and tries to minimize error.
For example, it would convert a RX-RZ-RX sequence to a RZ-RX-RZ sequence over the same basis, if RZ is cheaper. Previously the cost heuristic was something simple like the length of the sequence.

import math
from qiskit.transpiler import PassManager, Target, InstructionProperties
from qiskit.circuit.library import RZGate, RXGate, SXGate, CXGate
from qiskit.circuit import QuantumCircuit, Parameter, Measure
from qiskit.transpiler.passes import Optimize1qGatesDecomposition

θ = Parameter('θ')

rx_props = {
    (0,) : InstructionProperties(duration=0.5e-8, error=.00025)
}

rz_props = {
    (0,): InstructionProperties(duration=0, error=0)
}

target = Target()
target.add_instruction(RZGate(θ), rz_props, name='rz')
target.add_instruction(RXGate(θ), rx_props, name='rx')

circuit = QuantumCircuit(1)
circuit.sx(0)
circuit.barrier()
circuit.u(.1, .2, .3, 0)
circuit.barrier()
circuit.u2(.2, .3, 0)
circuit.barrier()
circuit.rx(.2, 0)
circuit.rz(.3, 0)
circuit.rx(.2, 0)
circuit.barrier()
circuit.u3(0, 0, math.pi/4, 0)
circuit.barrier()
circuit.rz(0, 0)
circuit.barrier()
circuit.rx(.75, 0)
circuit.rx(-.75, 0)
pm = PassManager(Optimize1qGatesDecomposition(target=target))
new_circuit = pm.run(circuit)

print('before:')
print(circuit)

print('after:')
print(new_circuit)

before:
   ┌────┐ ░ ┌────────────────┐ ░ ┌─────────────┐ ░ ┌─────────┐┌─────────┐┌─────────┐ ░ ┌─────────────┐ ░ ┌───────┐ ░ ┌──────────┐┌───────────┐
q: ┤ √X ├─░─┤ U(0.1,0.2,0.3) ├─░─┤ U2(0.2,0.3) ├─░─┤ Rx(0.2) ├┤ Rz(0.3) ├┤ Rx(0.2) ├─░─┤ U3(0,0,π/4) ├─░─┤ Rz(0) ├─░─┤ Rx(0.75) ├┤ Rx(-0.75) ├
   └────┘ ░ └────────────────┘ ░ └─────────────┘ ░ └─────────┘└─────────┘└─────────┘ ░ └─────────────┘ ░ └───────┘ ░ └──────────┘└───────────┘

after:
global phase: 1.6781
   ┌─────────┐ ░ ┌─────────────┐┌─────────┐┌────────────┐ ░ ┌─────────────┐┌─────────┐┌────────────┐ ░ ┌───────────┐┌─────────────┐┌───────────┐ ░ ┌─────────┐ ░  ░ 
q: ┤ Rx(π/2) ├─░─┤ Rz(-1.2708) ├┤ Rx(0.1) ├┤ Rz(1.7708) ├─░─┤ Rz(-1.2708) ├┤ Rx(π/2) ├┤ Rz(1.7708) ├─░─┤ Rz(0.153) ├┤ Rx(0.39545) ├┤ Rz(0.153) ├─░─┤ Rz(π/4) ├─░──░─
   └─────────┘ ░ └─────────────┘└─────────┘└────────────┘ ░ └─────────────┘└─────────┘└────────────┘ ░ └───────────┘└─────────────┘└───────────┘ ░ └─────────┘ ░  ░ 

[qiskit-bot]

Thank you for opening a new pull request.

Before your PR can be merged it will first need to pass continuous integration tests and be reviewed. Sometimes the review process can be slow, so please be patient.

While you're waiting, please feel free to review other open PRs. While only a subset of people are authorized to approve pull requests for merging, everyone is encouraged to review open pull requests. Doing reviews helps reduce the burden on the core team and helps make the project's code better for everyone.

One or more of the the following people are requested to review this:

• @Qiskit/terra-core

[coveralls]

Pull Request Test Coverage Report for Build 3492084298

• 62 of 62 (100.0%) changed or added relevant lines in 4 files are covered.
• 1 unchanged line in 1 file lost coverage.
• Overall coverage increased (+0.003%) to 84.579%

---

Files with Coverage Reduction	New Missed Lines	%
qiskit/transpiler/passes/optimization/optimize_1q_decomposition.py	1	98.72%

Totals
Change from base Build 3491436652:	0.003%
Covered Lines:	62511
Relevant Lines:	73908

---

💛 - Coveralls

[t-imamichi]

I'm assigned as reviewer, but I'm not familiar with the transpiler pass. @itoko -san may be more appropriate.

[mtreinish]

Overall this LGTM, thanks for doing this. I especially like that we're factoring the error rates into how we selection a decomposition now. I have a couple of small inline questions but besides those minor details I think this should be good to merge.

[mtreinish]

Thanks for the updates, I think this mostly LGTM now, I just caught one potential performance issue inline (the same one in both passes). Once that's fixed I think I'll be good to approve and merge this.

[mtreinish]

LGTM, thanks for the quick update.