Wavefunctions

zkit reads two kinds of wavefunction outputs:

Time-series snapshots from td/wfs_<input>.h5
Static eigenstates from static/EigenData_<input>.h5

Both contain IGA B-spline coefficients, not real-space arrays. zkit evaluates the basis with its bundled igakit.igalib.bsp compatibility layer so you can plot or export on a regular grid without installing a separate igakit package.

Read snapshots

from zkit.io.wavefunction import read_wfs

wfs = read_wfs("td/wfs_h2p.inp.h5")
print("snapshots:", wfs.n_snapshots)
print("data shape:", wfs.data.shape)
print("dimension:", wfs.dimension)

wfs.data is (n_snapshots, n_dof) complex coefficients in the IGA tensor order used by PetIGA.

Reconstruct a snapshot

from zkit.io.wfs_field import reconstruct_wfs

res = reconstruct_wfs(
    "td/wfs_h2p.inp.h5",
    step=0,
    npoints=240,
    axes=None,
)

print(res["dim"])
print(res["psi"].shape)
print(res["axes"][0].min(), res["axes"][0].max())

Returned keys:

dim
step
axes
psi
Re
Im
abs2
SplineDegree
nfuncs
knots

Reconstruct a static eigenstate

from zkit.io.eigen import read_eigen

eigen = read_eigen("static/EigenData_h2p.inp.h5")
res = eigen.reconstruct(istate=0, npoints=240)
print(res["psi"].shape)

Exact norm via Gauss quadrature

from zkit.io.wfs_field import compute_wfs_norm

norm = compute_wfs_norm("td/wfs_h2p.inp.h5", step=0, nq=5)
print(norm)

nq=5 is exact for spline degree <= 9. Higher degrees need larger nq.

VTK export

from zkit.io.wfs_field import reconstruct_wfs, wfs_to_vtk

res = reconstruct_wfs("td/wfs_h2p.inp.h5", step=0, npoints=120)
wfs_to_vtk(res, "figures/wfs.vts")

Supported extensions:

.vts StructuredGrid
.vtu UnstructuredGrid
.vtk legacy StructuredGrid

Three-dimensional reconstruction returns a tensor with shape (nx, ny, nz). For a quick visual check, the repository’s small 3D example plots the central z=0 slice of |psi|^2.