qomputing 0.1.6

Quantum computing state vector simulator with linear XEB benchmarking

Qomputing Simulator

A lightweight, pure Python state-vector simulator with linear cross-entropy benchmarking (XEB). For contributors and users who install from source.

---

Install

From the project root:

git clone <this-repo-url>
cd simulator   # or your clone directory
python3 -m venv .venv
source .venv/bin/activate   # Windows: .venv\Scripts\activate
pip install -e ".[dev]"

• [dev] adds pytest. For a normal (non-editable) install: pip install .

Without cloning (if the package is published):

pip install qomputing
# Optional: pip install qomputing[dev]

---

Usage

CLI

# Random circuit + XEB
qomputing-sim random-circuit --qubits 3 --depth 5 --shots 1000

# Simulate from JSON
qomputing-sim simulate --circuit path/to/circuit.json --shots 512 --seed 42

Optional: --max-qubits, --max-shots to cap resources; --single-qubit-gates, --two-qubit-gates, --multi-qubit-gates to override default gate sets.

Library

Backend API:

from qomputing import QomputingSimulator, QuantumCircuit

backend = QomputingSimulator.get_backend("state_vector")
qc = QuantumCircuit(4, 4)
qc.h(0).cx(0, 1).measure(range(3), range(3))
result = backend.run(qc, shots=1024)
print(result.get_counts())

Direct run:

from qomputing import run, QuantumCircuit, run_xeb, random_circuit

qc = QuantumCircuit(2)
qc.h(0).cx(0, 1)
result = run(qc, shots=1000, seed=42)
print(result.final_state, result.probabilities, result.counts)

qc = random_circuit(num_qubits=3, depth=5, seed=7)
xeb = run_xeb(qc, shots=1000, seed=7)
print(xeb.fidelity, xeb.sample_probabilities)

Rotation gates use qubit first, then angle: e.g. qc.rz(qubit, theta), qc.ry(qubit, theta).
Measure takes sequences: e.g. qc.measure([0], [0]) for one qubit; qc.measure_all() if num_clbits >= num_qubits.

---

Repository layout

qomputing/
├── circuit.py           # QuantumCircuit, gates, measure, JSON (de)serialization
├── circuit_builders.py   # random_circuit for XEB
├── gates/                # Single-, two-, multi-qubit gate implementations
├── engine/               # State vector core, measurements, registry
├── linalg.py             # Tensor / linear algebra helpers
├── cli.py                # qomputing-sim entry point
├── xeb.py                # Linear XEB fidelity
├── backend.py            # QomputingSimulator, Result
├── run.py                # run(), load_circuit(), run_xeb()
├── visualisation*.py     # Circuit drawer and Bloch (optional)
tools/
├── cirq_comparison.py    # Parity tests vs Cirq
tests/                    # Pytest: parity, limits, XEB, drawer, bugs

---

Circuit JSON format

{
  "num_qubits": 2,
  "gates": [
    {"name": "h", "targets": [0]},
    {"name": "cx", "controls": [0], "targets": [1]},
    {"name": "rz", "targets": [0], "params": {"theta": 1.5708}}
  ]
}

Supported gates: single-qubit id, x, y, z, h, s, sdg, t, tdg, sx, sxdg, rx, ry, rz, u1, u2, u3; two-qubit cx, cy, cz, cp, csx, swap, iswap, sqrtiswap, rxx, ryy, rzz; multi-qubit ccx, ccz, cswap.

---

Testing

pytest

Parity vs Cirq (optional):

export PYTHONPATH="$PWD"
python tools/cirq_comparison.py --min-qubits 1 --max-qubits 5 --depths 3 5 --circuits-per-config 3 --shots 256 --seed 7

---

Development

• Run pytest before committing.
• Use CONTRIBUTING.md for contribution guidelines (if present).
• Build package: python -m build (creates dist/). For publishing to PyPI, use ./scripts/publish-to-pypi.sh (maintainer only).

---

License

MIT