torchqc 1.1.0

Quantum simulation library based on PyTorch

TorchQC: Quantum Dynamics and Machine Learning

Description

TorchQC is a Python library that is based on the PyTorch deep learning library. Quantum operations and dynamics are entirely implemented using the PyTorch tensor mechanism. This makes all generated data ready for use in Deep Learning models.

Contributors

Dimitris Koutromanos Dionisis Stefanatos Emmanuel Paspalakis

Repository structure

• examples: a folder containing jupyter notebooks demonstrating how the library can be used

  • composite_system_dynamics.ipynb
  • correlation_functions.ipynb
  • coupled_qubits_plus_resonator.ipynb
  • coupled_qubits_resonator_BBNN_extra_costs.ipynb
  • coupled_qubits_resonator_BBNN.ipynb
  • coupled_spins_3lvl_system.ipynb
  • demonstrations.ipynb
  • Dynamical_blockade_coupled_bosonic_modes.ipynb
  • Dynamical_blockade_single_mode_bosonic_system.ipynb
  • Dynamical_blockade_single_mode_gaussian_pulses.ipynb
  • emission_spectum.ipynb
  • Jaynes_Cummings_model_cat_state.ipynb
  • Jaynes_Cummings_model_coherent_state.ipynb
  • Jaynes_Cummings_model.ipynb
  • Jaynes_Cummings_model_photon_number_parity_oscillations.ipynb
  • lambda_3lvl_system_BBNN.ipynb
  • lambda_3lvl_system_STIRAP.ipynb
  • Lossy_Jaynes_Cummings_model_coherent_state.ipynb
  • Lossy_Jaynes_Cummings_model.ipynb
  • optomechanical_system.ipynb
  • Quantum_Rabi_model.ipynb
  • qubit_rabi_oscillations.ipynb
  • single_bosonic_mode.ipynb
  • single_qubit_dynamics.ipynb
  • single_qubit_markovian_master_equation_fock_liouville.ipynb
  • single_qubit_markovian_master_equation_num_methods.ipynb
  • steadystate.ipynb
  • Two_operator_two_time_correlation_function.ipynb

• LICENSE: license file

• README.md: Readme Markdown file

• requirements.txt: file with Python package requirements for running the library

• torchqc: main library directory with the core source code

  • adams_bashforth_methods.py: Adams numerical methods for solving ordinary differential equations
  • common_functions.py: contains common useful functions
  • common_matrices.py: matrix representations of common quantum operators
  • correlation.py: methods to calculate correlation functions
  • dynamics.py: methods to solve quantum dynamics (i.e. Schrodinger equation, Lindblad master equation)
  • fock_liouville.py: methods to calculate operators useful for quantum dynamics in Fock-Liouville space
  • mappings.py: Lie algebra operations (commutator, anticommutator)
  • markovian.py: method to apply Lindblad dissipative operator
  • operators.py: Operator / DynamicOperator Python classes that represent arbitrary quantum operators (density matrix, Hamiltonian)
  • runge_kutta_methods.py: Runge-Kutta numerical methods for solving ordinary differential equations
  • states.py: QuamtumState class representing quantum state vector
  • tensor_product.py: methods that calculate tensor products (Kronecker products) of states or operators

Installation instructions

Alternative 1 (From PyPI)

• Install package directly from PyPI

pip install torchqc

Alternative 2 (From source code)

• Clone GitHub repository

git clone https://github.com/qoptics-qtech/torchqc.git

• Setup Python virtual environment (Python 3 required)

python -m venv my_env

• Activate venv

source my_env/bin/activate

• Install required packages

pip install -r requirements.txt

• Install package using src code

pip install .

Examples execution

The examples folder contains several examples that can be added into three main categories

• Closed and Open Quantum system simulations
• Quantum optimal control with machine learning
• Quantum optics

All examples are implemented in jupyter notebooks and can be executed by any program that can execute a jupyter notebook, such as Visual Code Studio

Usage instructions

To use the library, after it is successfully installed, it is required to be added to the program scope using commands such as:

import torchqc

Or directly including the submodules of interest:

from torchqc.states import QuantumState

After the needed object is in scope, one can use it to perform quantum operations. The following code defines the ground state of a qubit system:

qubit_state = QuantumState.basis(2)[0]