module moplayground.utils.plotting
function plot_mo_progress
plot_mo_progress(
num_steps: int,
metrics: dict,
training_data: moplayground.utils.plotting.MOTrainingPlottingInfo,
save_dir: pathlib.Path,
run: wandb.sdk.wandb_run.Run = None
)
function default_coloring
default_coloring(tradeoff)
function plot_pareto
plot_pareto(
ax: matplotlib.axes._axes.Axes,
pareto: numpy.ndarray,
colors: str | numpy.ndarray = None,
objective: list[str] = None,
connect: bool = False,
show_dominated: bool = True,
nondominated_alpha: float = 1.0,
dominated_alpha: float = 1.0,
nondominated_s: int = 20,
dominated_s: int = 8,
outline_nondominated: float = 0.0,
label: str = None,
set_lims: bool = True,
label_fontsize: int = 16,
special_idxs: numpy.ndarray = None,
special_marker: str = '*',
connect_line_color: str | numpy.ndarray = 'black',
**plot_kwargs
)
Plot a pareto frontier.
colors is a single matplotlib color or per-point color. RGB(A) label names the front in the legend; set_lims zooms to the nondominated front.
function plot_sequential_paretos
plot_sequential_paretos(
ax_titles: list[str],
paretos: numpy.ndarray,
directives: numpy.ndarray = None,
objectives: list[str] = None
)
function plot_sequential_hypervolume
plot_sequential_hypervolume(
iterations: list[int] | numpy.ndarray,
paretos: numpy.ndarray
)
class MOTrainingPlottingInfo
MOTrainingPlottingInfo(start_time: float, times: list =
method MOTrainingPlottingInfo.__init__
__init__(
start_time: float,
times: list = <factory>,
iterations: list = <factory>,
paretos: list = <factory>,
directives: list = <factory>,
labels: list = <factory>,
hypervolumes: list = <factory>,
sparsities: list = <factory>
) → None
method MOTrainingPlottingInfo.save
save(save_dir, create_time=True)