cocopp.readalign

module documentation

Helper routines to read in data files.

The terms horizontal and vertical below refer to the horizontal (fixed-target) and vertical (fixed-budget) views. When considering convergence graphs of function values over times, we can view it as:

  • costs for different fixed targets represented by horizontal cuts.
  • function values for different fixed budgets represented by vertical cuts.

COCO collects experimental data with respect to these two complementary views. This module provides data structures and methods for dealing with the experimental data.

Class ArrayMultiReader Class of aligned data arrays to be aligned together.
Class HArrayMultiReader Wrapper class of aligned data arrays to be aligned horizontally.
Class HMultiReader List of data arrays to be aligned horizontally.
Class MultiReader List of data arrays to be aligned.
Class VArrayMultiReader Wrapper class of aligned data arrays to be aligned vertically.
Class VArrayMultiReaderNew Wrapper class of aligned data arrays to be aligned vertically.
Class VMultiReader List of data arrays to be aligned vertically.
Function align_data Aligns the data from a list of data arrays.
Function alignArrayData Aligns the data from a list of aligned arrays.
Function is_close Undocumented
Function openfile kwargs are passed to open
Function split Split a list of data files into arrays corresponding to data sets. The Boolean list idx_to_load is thereby indicating whether a given part of the split is to be considered or not if None, all instances are considered.
def align_data(data, idx_evals, idx_funvals, rewind_reader=False):

Aligns the data from a list of data arrays.

This method returns an array for which the alignment value is the first column and the aligned values are in subsequent columns.

def alignArrayData(data):

Aligns the data from a list of aligned arrays.

This method returns an array for which the alignment value is the first column and the aligned values are in subsequent columns.

def is_close(a, b, rel_tol=1e-09, abs_tol=0.0):

Undocumented

def openfile(filePath, **kwargs):

kwargs are passed to open

def split(dataFiles, idx_to_load=None, dim=None):

Split a list of data files into arrays corresponding to data sets. The Boolean list idx_to_load is thereby indicating whether a given part of the split is to be considered or not if None, all instances are considered.