TuneQ

A production-ready library for measurement-error mitigation and circuit optimization in NISQ computing.

---

Table of Contents

1. Overview
2. Key Features
3. Installation
   • Using Poetry
   • Using Requirements Files (pip)
4. Usage
   • Measurement-Error Mitigation
   • Circuit Optimization
   • Example: Qiskit Integration
   • Example: Cirq Integration
5. Contributing
6. License

---

Overview

TuneQ is designed to simplify two core tasks in NISQ (Noisy Intermediate-Scale Quantum) computing:

1. Measurement-Error Mitigation: Build calibration circuits, invert the resulting calibration matrices, and automatically correct readout errors in multi-qubit systems.
2. Lightweight Circuit Optimization: Detect whether you’re using Qiskit or Cirq and invoke the corresponding transpiler or optimizer with minimal fuss.

TuneQ helps researchers, developers, and quantum enthusiasts spend more time focusing on their algorithms and less time writing boilerplate code for calibration, error mitigation, or circuit optimization.

---

Key Features

• Single- or Multi-Qubit Readout Calibration:

  • Tensor-product approach for multi-qubit calibration.
  • Automatically builds and runs calibration circuits, then creates an error-mitigation matrix.
  • Applies matrix inversion to correct measurement results.

• One-Line Circuit Optimization:

  • Auto-detects whether you’re using Qiskit or Cirq.
  • Leverages native transpile/optimization passes from each framework.
  • Bypasses complex pass-manager setups in Qiskit or advanced transformations in Cirq.

• Flexible & Framework-Agnostic:

  • No forced login to specific quantum cloud providers.
  • Users provide their own function to run circuits on any backend (real device or simulator).

• Error Handling & Diagnostics:

  • Custom exceptions (CalibrationError, OptimizationError) provide clear feedback.
  • Handles ill-conditioned calibration matrices gracefully.

---

Installation

TuneQ supports Python 3.10+. You can install via Poetry or generate a requirements.txt for pip.

Using Poetry

1. Clone the repo:

   git clone https://github.com/cheikhmc/tuneq.git
   cd tuneq

2. Install with poetry:

   poetry install

   By default, TuneQ installs only the core dependencies (e.g numpy). To enable Qiskit or Cirq integrations:

   poetry install --extras "qiskit"
   poetry install --extras "cirq"

3. Run Tests

   poetry run pytest

Usage

Measurement-Error Mitigation

from tuneq import mitigate_measurement_errors

with mitigate_measurement_errors(
    num_qubits=2,
    run_circuits=my_runner_function,
    shots=1024
) as mitigator:
    corrected_counts = mitigator.run(my_main_circuit)

Circuit Optimization

from tuneq import optimize_circuit_for

optimized_circuit = optimize_circuit_for(my_circuit, optimization_level=2)

Qiskit Example

from qiskit import QuantumCircuit, Aer, transpile
from tuneq import mitigate_measurement_errors, optimize_circuit_for

qc = QuantumCircuit(2)
qc.h(0)
qc.cx(0, 1)
qc.measure_all()

def qiskit_runner(circuits, shots=1024):
    simulator = Aer.get_backend('qasm_simulator')
    return [simulator.run(transpile(c, simulator), shots=shots).result().get_counts() for c in circuits]

with mitigate_measurement_errors(num_qubits=2, run_circuits=qiskit_runner, shots=1024) as mitigator:
    corrected_counts = mitigator.run(qc)

optimized_qc = optimize_circuit_for(qc, optimization_level=3)

Contributing

We welcome contributions! Feel free to:

• Open an issue to report bugs or suggest features.
• Submit a pull request for fixes or new features.
• Join discussions to collaborate on improvements.

---

License

This project is licensed under the MIT License.