history

Name	Description
`plot_history_metrics`	Plot training history metrics

Name

Description

plot_history_metrics

Plot training history metrics

plot_history_metrics(
    history: dict[str, list[float]],
    metrics: list[str],
    save_path: Path | None = None,
    include_val: bool = True,
    figsize: tuple[int, int] = (9, 5),
    colors: tuple[str | tuple[str, str]] = ('blue', 'orange'),
    stack: bool = False,
    title: str | None = None,
    **kwargs={},
) -> tuple[plt.Figure, plt.Axes]

Plot training history metrics returned by model.fit.

Example:

history = dict(
    loss=[0.1, 0.2, 0.3, 0.4],
    accuracy=[0.9, 0.8, 0.7, 0.6],
    val_loss=[0.1, 0.2, 0.3, 0.4],
    val_accuracy=[0.9, 0.8, 0.7, 0.6],
)

import helia_edge as helia

fig, ax = helia.plotting.plot_history_metrics(
    history,
    metrics=["loss", "accuracy"],
    include_val=True,
    stack=False,
)

Parameters

Parameters of plot_history_metrics
Name	Type	Default	Description
`history`	`dict[str, list[float]]`	Required	Training history
`metrics`	`list[str]`	Required	Metrics to plot
`save_path`	`Path \| None`	None	Path to save plot. Defaults to None.
`include_val`	`bool`	True	Include validation metrics. Defaults to True.
`figsize`	`tuple[int, int]`	(9, 5)	Figure size. Defaults to (9, 5).
`colors`	`tuple[str \| tuple[str, str]]`	('blue', 'orange')	Colors for train and val. Defaults to ("blue", "orange").
`stack`	`bool`	False	Stack metrics. Defaults to False.
`title`	`str \| None`	None	Title for plot. Defaults

Returns

Returns of plot_history_metrics
Type	Description
`tuple[plt.Figure, plt.Axes]`	tuple[plt.Figure, plt.Axes]: Figure and axes handles

history

Training History Plotting API