Realistic post-processing campaign

This example follows a complete TDSEZ analysis workflow for a 1D hydrogen-like run. The same pattern applies to 2D and 3D with minor dimension-dependent adjustments.

Run setup

Assume the TDSEZ binary produced:

/data/h2p/
  td/
    TimeEvolutionData_h2p.inp.h5
    wfs_h2p.inp.h5
  static/
    EigenData_h2p.inp.h5

1. Summarize the run

zkit summary h2p.inp --run-dir /data/h2p

Confirm the expected dimension, step count, and state count before loading data.

2. Load everything in Python

from zkit.simulation import Run

run = Run("/data/h2p", "h2p.inp")
print("dimension:", run.dim)
print("input:", run.meta.input_file)
print("time step:", run.meta.time_step)
print("final time:", run.meta.final_time)
print("n_steps:", run.evolution.n_steps)
print("states:", run.spectrum.n_states)
print("wfs snapshots:", run.wavefunctions.n_snapshots)

3. Inspect the spectrum

import numpy as np

E = run.spectrum.values
print("ground state:", E[0])
print("first excited:", E[1])
print("spacing:", np.diff(E)[:5])

4. Plot energy over time

import matplotlib.pyplot as plt

time = run.time
energies = run.energies

plt.figure(figsize=(8, 4))
plt.plot(time, energies[:, 4], label="total energy")
plt.plot(time, energies[:, 1], label="kinetic", alpha=0.6)
plt.plot(time, energies[:, 2], label="potential", alpha=0.6)
plt.xlabel("time (a.u.)")
plt.ylabel("energy (a.u.)")
plt.legend()
plt.tight_layout()
plt.savefig("figures/h2p_energy.png", dpi=150)

5. Plot dipole and acceleration

dipoles = run.dipoles
time = dipoles[:, 0]
dipole_x = dipoles[:, 2]
accel_x = dipoles[:, 3]

plt.figure(figsize=(8, 4))
plt.plot(time, dipole_x, label="dipole")
plt.plot(time, accel_x, label="acceleration", alpha=0.7)
plt.xlabel("time (a.u.)")
plt.legend()
plt.tight_layout()
plt.savefig("figures/h2p_dipole.png", dpi=150)

6. Inspect populations

pop = run.populations
print(pop.shape)
print(pop[0, 1:])
print(pop[-1, 1:])

7. Reconstruct and plot wavefunctions

from zkit.viz import plot_wavefunction

plot_wavefunction(
    "/data/h2p/td/wfs_h2p.inp.h5",
    step=0,
    outdir="figures",
    npoints=240,
    dpi=150,
    vtk="figures/wfs_step0.vts",
)

8. Reconstruct selected eigenstates

from zkit.io.eigen import read_eigen

eigen = read_eigen("/data/h2p/static/EigenData_h2p.inp.h5")
for i in range(3):
    res = eigen.reconstruct(istate=i, npoints=300)
    x = res["axes"][0]
    psi = res["psi"]
    np.savez(f"figures/eigen_{i}.npz", x=x, psi=psi)

9. Check norm conservation

from zkit.io.wfs_field import compute_wfs_norm
from pathlib import Path

wfs_path = Path("/data/h2p/td/wfs_h2p.inp.h5")
for step in [0, 50, 100, 150]:
    print(step, compute_wfs_norm(wfs_path, step=step, nq=5))

10. Export selected snapshots to VTK for ParaView

from zkit.io.wfs_field import reconstruct_wfs, wfs_to_vtk
from pathlib import Path

outdir = Path("figures/vtk")
outdir.mkdir(exist_ok=True)

for step in [0, 50, 100]:
    res = reconstruct_wfs(wfs_path, step=step, npoints=120)
    wfs_to_vtk(res, outdir / f"wfs_step{step}.vts")

Resulting artifacts

figures/h2p_energy.png
figures/h2p_dipole.png
figures/wfs_step0.png
figures/wfs_step0.vts
figures/eigen_0.npz, eigen_1.npz, eigen_2.npz
figures/vtk/wfs_step{0,50,100}.vts