lumopt2.utils.panels

Reusable panels for lumopt2.utils.graphical_visualizer.GraphicalVisualizer.

A panel is a small, self-contained widget that owns one matplotlib subplot and knows how to render its slice of the optimization figure. The GraphicalVisualizer lays panels out in a grid and dispatches the standard setup / update / on_optimization_end lifecycle calls to each one, so users can compose figures from any combination of built-in panels (FOM, gradient norm, geometry, monitor data) and custom Panel subclasses they write themselves.

Classes

Panel([title])	Abstract base class for a single subplot in a GraphicalVisualizer.
PanelState(iteration, iterations, ...)	Immutable per-update snapshot handed to every panel's update call.
FomPanel([title, show_best, log_y, color, ...])	Live plot of the figure of merit vs.
GeometryPanel([title, parametrization, ...])	Live plot of the current geometry on top of the initial baseline.
GradientNormPanel([title, auto_log, color, ...])	Live plot of the gradient L2 norm vs.
MonitorPanel([title, monitor_name, ...])	Live plot of an FDTD monitor result (field component, S-parameter, ...).