cocopp.compall.pprldmany

module documentation

Generates figure of the bootstrap distribution of (function) evaluations.

The main method in this module generates figures of Empirical Cumulative Distribution Functions of the bootstrap distribution of the (function) evaluations needed to reach a target divided by the dimension for many algorithms.

The outputs show the ECDFs of the running times of the simulated runs divided by dimension for 50 different targets logarithmically uniformly distributed in [1e−8, 1e2]. The crosses (×) give the median number of function evaluations of unsuccessful runs divided by dimension.

Example

Function	`all_single_functions`	Undocumented
Function	`beautify`	Customize figure presentation.
Function	`main`	Generates a figure showing the performance of algorithms.
Function	`plot`	This function is obsolete? Generates a graph of the run length distribution of an algorithm.
Function	`plotdata`	Draw a normalized ECDF. What means normalized?
Function	`plotLegend`	Display right-side legend.
Function	`plt_plot`	Undocumented
Function	`text_infigure_if_constraints`	to be displayed in the figure corner
Constant	`EDA`	Undocumented
Constant	`GA`	Undocumented
Constant	`MC`	Undocumented
Constant	`TAO`	Undocumented
Variable	`annotation_line_end_relative`	Undocumented
Variable	`annotation_space_end_relative`	Undocumented
Variable	`best`	Undocumented
Variable	`best2`	Undocumented
Variable	`bestbest`	Undocumented
Variable	`bestnoisy`	Undocumented
Variable	`budget_cross_always`	Undocumented
Variable	`classics`	Undocumented
Variable	`close_figure`	Undocumented
Variable	`displaybest`	Undocumented
Variable	`divide_by_dimension`	Undocumented
Variable	`eseda`	Undocumented
Variable	`ESs`	Undocumented
Variable	`fmulti`	Undocumented
Variable	`funi`	Undocumented
Variable	`funilipschitz`	Undocumented
Variable	`funisep`	Undocumented
Variable	`label_fontsize`	Undocumented
Variable	`max_evals_marker_format`	Undocumented
Variable	`max_evals_percentile`	Undocumented
Variable	`max_evals_single_marker_format`	Undocumented
Variable	`nbperdecade`	Undocumented
Variable	`nikos`	Undocumented
Variable	`nikos40D`	Undocumented
Variable	`petr`	Undocumented
Variable	`PlotType`	Undocumented
Variable	`save_figure`	Undocumented
Variable	`save_zoom`	Undocumented
Variable	`show_algorithms`	Undocumented
Variable	`size_correction_from_n_foreground`	Undocumented
Variable	`TAOp`	Undocumented
Variable	`third`	Undocumented
Variable	`title_fontsize`	Undocumented
Variable	`xticks_fontsize`	Undocumented
Variable	`yticks_fontsize`	Undocumented

def all_single_functions(dict_alg, is_single_algorithm, sorted_algs=None, output_dir='.', parent_html_file_name=None, settings=genericsettings): ¶

Undocumented

def beautify(): ¶

Customize figure presentation.

def main(dictAlg, order=None, outputdir='.', info='default', dimension=None, parentHtmlFileName=None, plotType=PlotType.ALG, settings=genericsettings): ¶

Generates a figure showing the performance of algorithms.

From a dictionary of DataSetList sorted by algorithms, generates the cumulative distribution function of the bootstrap distribution of evaluations for algorithms on multiple functions for multiple targets altogether.

Parameters
dictAlg	Undocumented
order	Undocumented
outputdir	Undocumented
info	Undocumented
dimension	Undocumented
parentHtmlFileName	Undocumented
plotType	Undocumented
settings	Undocumented
dict dictAlg	dictionary of `DataSetList` instances one instance is equivalent to one algorithm,
list targets	target function values
list order	sorted list of keys to dictAlg for plotting order
str outputdir	output directory
str info	output file name suffix
str parentHtmlFileName	defines the parent html page

def plot(dsList, targets=None, craftingeffort=0.0, **kwargs): ¶

This function is obsolete? Generates a graph of the run length distribution of an algorithm.

We display the empirical cumulative distribution function ECDF of the bootstrapped distribution of the runlength for an algorithm (in number of function evaluations) to reach the target functions value targets.

Parameters
dsList	Undocumented
targets	Undocumented
craftingeffort	Undocumented
DataSetList dsList	data set for one algorithm
seq targets	target function values
float crafting effort	the data will be multiplied by the exponential of this value
dict kwargs	additional parameters provided to plot function.
**kwargs	Undocumented
Returns
handles

def plotdata(data, maxval=None, maxevals=None, CrE=0.0, maxevals2=None, **kwargs): ¶

Draw a normalized ECDF. What means normalized?

Parameters
data	Undocumented
maxval	Undocumented
maxevals	Undocumented
CrE	Undocumented
maxevals2	a single value or values to be plotted as median(maxevals2) with the same marker as maxevals
seq data	data set, a 1-D ndarray of runlengths
float maxval	right-most value to be displayed, will use the largest non-inf, non-nan value in data if not provided
seq maxevals	if provided, will plot the median of this sequence as a single cross marker
float CrE	Crafting effort the data will be multiplied by the exponential of this value
**kwargs	optional arguments provided to plot function.

def plotLegend(handles, maxval): ¶

Display right-side legend.

The figure is stopped at maxval (upper x-bound), and the graphs in the figure are prolonged with straight lines to the right to connect with labels of the graphs (uniformly spread out vertically). The order of the graphs at the upper x-bound line give the order of the labels, in case of ties, the best is the graph for which the x-value of the first step (from the right) is smallest.

The annotation string is stripped from preceeding pathnames.

Parameters
handles	Undocumented
maxval	Undocumented
float maxval	rightmost x boundary
Returns
list of (ordered) labels and handles.

def plt_plot(*args, **kwargs): ¶

Undocumented

def text_infigure_if_constraints(): ¶

to be displayed in the figure corner

TODO: is a method with no arguments because if made a variable, an error is raised as it is computed before testbedsettings.current_testbed is instantiated in some import

EDA: tuple[str, ...] = ¶

Undocumented

Value

('BIPOP-CMA-ES',
 '(1+1)-CMA-ES',
 'VNS (Garcia)',
 'EDA-PSO',
 'IPOP-SEP-CMA-ES',
 'AMaLGaM IDEA',
 'iAMaLGaM IDEA',
...

GA: tuple[str, ...] = ¶

Undocumented

Value

('DE-PSO',
 '(1+1)-ES',
 'PSO_Bounds',
 'DASA',
 'G3-PCX',
 'simple GA',
 'POEMS',
...

MC: tuple[str, ...] = ¶

Undocumented

Value

('Monte Carlo')

TAO: tuple[str, ...] = ¶

Undocumented

Value

('BFGS',
 'NELDER (Han)',
 'NEWUOA',
 'full NEWUOA',
 'BIPOP-CMA-ES',
 'IPOP-SEP-CMA-ES',
 '(1+1)-CMA-ES',
...

annotation_line_end_relative: float = ¶

Undocumented

annotation_space_end_relative: float = ¶

Undocumented

best: tuple[str, ...] = ¶

Undocumented

best2: tuple[str, ...] = ¶

Undocumented

bestbest: tuple[str, ...] = ¶

Undocumented

bestnoisy: tuple = ¶

Undocumented

budget_cross_always: bool = ¶

Undocumented

classics: tuple[str, ...] = ¶

Undocumented

close_figure: bool = ¶

Undocumented

displaybest: bool = ¶

Undocumented

divide_by_dimension: bool = ¶

Undocumented

eseda: tuple[str, ...] = ¶

Undocumented

ESs: tuple[str, ...] = ¶

Undocumented

fmulti = ¶

Undocumented

funi = ¶

Undocumented

funilipschitz = ¶

Undocumented

funisep: list[int] = ¶

Undocumented

label_fontsize: int = ¶

Undocumented

max_evals_marker_format: list = ¶

Undocumented

max_evals_percentile: int = ¶

Undocumented

max_evals_single_marker_format: list = ¶

Undocumented

nbperdecade: int = ¶

Undocumented

nikos: tuple[str, ...] = ¶

Undocumented

nikos40D: tuple[str, ...] = ¶

Undocumented

petr: tuple[str, ...] = ¶

Undocumented

PlotType = ¶

Undocumented

save_figure: bool = ¶

Undocumented

save_zoom: bool = ¶

Undocumented

show_algorithms: tuple = ¶

Undocumented

size_correction_from_n_foreground: int = ¶

Undocumented

TAOp = ¶

Undocumented

third: tuple[str, ...] = ¶

Undocumented

title_fontsize: int = ¶

Undocumented

xticks_fontsize: int = ¶

Undocumented

yticks_fontsize: int = ¶

Undocumented