Tutorials and learning paths
PyQED is broad, so the safest route is to start with one small, inspectable workflow and follow its linked guide, example, and test. The paths below do not imply that every combination of options is supported.
Quantum chemistry
Run the native RHF Five-minute quickstart.
Read Quantum Chemistry for molecule construction and solver families.
Read Backends and Integral Representations before selecting dense, packed, RI, or factorized electron-repulsion data.
Continue to MP2 and COMP2 or CASCI and CASSCF.
Inspect
examples/qchem/sa_casscf_factor.pyand its corresponding focused tests before scaling the problem.
GW and response calculations are separate advanced paths; begin with GW and BSE or TDDFT and Ehrenfest Dynamics and use their documented conventions.
Grid and wavepacket dynamics
Read the representation overview in Discrete Variable Representation.
Run
examples/dvr/fedvr_harmonic_oscillator.py. It prints computed eigenvalues beside the analytic harmonic-oscillator sequence.Compare grid families with
examples/dvr/fedvr_vs_sine_quartic.py.Move to
examples/dvr/gwp_sddvr_2d_independent_ho.pyonly after the one-dimensional smoke case is understood.
Nonadiabatic and geometric dynamics
Read Geometric Quantum Dynamics and Nonadiabatic Molecular Dynamics.
Start with
examples/namd/ehrenfest.pyorexamples/namd/ldrfg_avoided_crossing.py.Treat ab initio and sparse-grid scripts as research workflows: inspect their optional dependencies, cached inputs, grid convergence, and output paths before execution.
Open systems and spectroscopy
Read Open quantum dynamics and HEOM and structured baths for open-system conventions.
Use
examples/heom.pyas an entry point, then inspect focused tests for the exact solver path being used.For spectroscopy, read Nonlinear molecular spectroscopy and inspect
examples/signals/absorption.pybefore the largerexamples/2DES.pyworkflow.
Floquet and light–matter models
Read Floquet and Polariton, then inspect
examples/floquet/two_level_system.py and
examples/floquet/RiceMele.py. These are model-specific entry points;
verify units and gauge conventions before adapting them.
How to turn an example into research evidence
For every path:
pin the PyQED release or Git commit;
keep the exact input and random seed;
record units, basis/grid, solver tolerances, dependencies, and threads;
establish convergence with respect to the relevant numerical controls; and
use the Benchmarks and validation manifest for results presented as validation or performance evidence.