Skip to content

API reference

Full per-symbol autodoc is wired in at release (once the engines are pip-installable by tag). For now, the key public entry points — the authoritative reference is each engine's own docstrings:

dmipy-sim — github.com/dmrai-lab/dmipy-sim

  • simulate(n_walkers, diffusivity, waveform, geometry, T2=None, ...) — walker-ensemble signal.
  • simulate_cpmg(n_walkers, D, cpmg_waveform, geometry, T2=None) — full CPMG echo train from one walk.
  • simulate_mixture(compartments, waveform) — fraction-weighted multi-compartment signal.
  • Encodings: pgse, ogse, cpmg, ste, pte, trapezoidal_ogse; set_b, calc_b, calc_btensor.
  • Geometries: FreeDiffusion, Box1D, Sphere, Cylinder, Ellipsoid, PackedCylinders, PackedSpheres, MyelinatedCylinder, PackedMyelinatedCylinders.

dmipy-fit — github.com/dmrai-lab/dmipy-fit

  • core.modeling_framework.MultiCompartmentModel — build + fit(scheme, data, solver=...).
  • signal_models.* — C1Stick, G1Ball, G2Zeppelin, sphere/plane/capped-cylinder.
  • signal_models.attenuation.OccupancyGatedModel + TransverseRelaxation, IntraPoreSurfaceRelaxivity, ExteriorSurfaceRelaxivity.
  • white_matter.build_white_matter_model(...) — decoupled diffusion-only canonical WM model.
  • white_matter.t2_spectrum_mwf(signal, echo_times, ...) — standard NNLS myelin-water fraction.
  • white_matter.surface.exterior_surface_to_volume(f_axon, gamma_shape, gamma_scale_outer_diameter).