Circuit Analysis

analyze_circuit() extracts routing- and telemetry-relevant features from a QuantSDK circuit — the same metrics QuantRouter uses for backend selection and that the cloud platform logs for ML training. It is a pure, read-only analysis of a circuit (no execution).

Usage

import quantsdk as qs

circuit = (
    qs.Circuit(3, name="ghz")
    .h(0)
    .cx(0, 1)
    .cx(0, 2)
    .measure_all()
)

features = qs.analyze_circuit(circuit)

print(features.qubit_count)   # 3
print(features.depth)         # circuit depth
print(features.cx_count)      # 2
print(features.algorithm_class)

Both names are top-level public API:

from quantsdk import analyze_circuit, CircuitFeatures

They also remain available via the router submodule (qs.router.analyze_circuit).

CircuitFeatures

A frozen dataclass returned by analyze_circuit().

Field	Type	Description
qubit_count	int	Number of qubits in the circuit.
gate_count	int	Total number of gates.
depth	int	Circuit depth (longest input→output path).
cx_count	int	Number of two-qubit (entangling) gates.
gate_types	frozenset[str]	Set of gate type names used.
single_qubit_gates	int	Count of single-qubit gates.
two_qubit_gates	int	Count of two-qubit gates.
three_qubit_gates	int	Count of three-qubit gates.
measurement_count	int	Number of measurement operations.
has_parameterized_gates	bool	Whether the circuit has rotation gates.
algorithm_class	AlgorithmClass	Detected algorithm class (e.g. entangling, variational).
connectivity	frozenset[tuple[int, int]]	Qubit pairs that interact via 2-qubit gates.

Because it is frozen, a CircuitFeatures instance is hashable and safe to cache or log.