rgpycrumbs.eon.plt_min

Plots minimization optimization trajectories.

Added in version 1.3.0.

Visualizes minimization runs using the generalized (s, d) reaction valley projection. Supports:

1. Energy Profile: Energy vs iteration.

2. 2D Optimization Landscape: Projects trajectory into (progress, deviation) coordinates relative to (initial, minimum).

3. Convergence Panel: Force norm and step size vs iteration.

Attributes

log
IRA_KMAX_DEFAULT
_OVERLAY_COLORS

Functions

main(job_dir, label, prefix, plot_type, project_path, ...)	Plot minimization trajectory visualization.
_plot_profile(trajs, labels, output, dpi)
_plot_landscape(trajs, labels, output, dpi, *, ...[, ...])
_plot_convergence(trajs, labels, output, dpi)

Module Contents

rgpycrumbs.eon.plt_min.log

rgpycrumbs.eon.plt_min.IRA_KMAX_DEFAULT = 1.8

rgpycrumbs.eon.plt_min.main(job_dir, label, prefix, plot_type, project_path, surface_type, ira_kmax, theme, plot_structures, strip_renderer, strip_spacing, strip_dividers, rotation, perspective_tilt, output, dpi, verbose)

Plot minimization trajectory visualization.

Use –job-dir multiple times to overlay trajectories from different optimizers (e.g. FIRE, LBFGS, SD) on the same landscape or profile.

rgpycrumbs.eon.plt_min._OVERLAY_COLORS = ['#004D40', '#FF655D', '#3F51B5', '#FF9800', '#9C27B0', '#009688']

rgpycrumbs.eon.plt_min._plot_profile(trajs, labels, output, dpi)

rgpycrumbs.eon.plt_min._plot_landscape(trajs, labels, output, dpi, *, project_path, surface_type, ira_kmax, cmap='viridis', plot_structures='none', strip_renderer='xyzrender', strip_spacing=1.5, strip_dividers=False, rotation='auto', perspective_tilt=0.0, theme=None)

rgpycrumbs.eon.plt_min._plot_convergence(trajs, labels, output, dpi)