package documentation

Experimentation module of the COCO - COmparing Continuous Optimizers - framework.

The module provides benchmark test beds in the Suite class and output data facilities in the Observer class.

See the documentation of the Suite class:

>>> import cocoex as ex
>>> help(ex.Suite)  # doctest: +ELLIPSIS
Help on class Suite...
>>> print(ex.known_suite_names)  # doctest: +ELLIPSIS
[...

A more complete example use case can be found in the example_experiment.py file.

Module exceptions Undocumented
Module solvers No module docstring; 1/1 function documented
Module utilities No module docstring; 8/8 functions, 5/5 classes documented
Module _version Undocumented

From __init__.py:

Class BatchScheduler Facilitate to run a benchmarking experiment in independent batches.
Class ExperimentRepeater Allow to automatically repeat an experiment based on budget and successes.
Class Observer Observer which can be "attached to" one or several problems, however not necessarily at the same time.
Class Problem Problem instances are usually generated using class Suite.
Class Suite Suite of benchmark problems.
Function default_observers return a map from suite names to default observer names.
Variable _default_observers Undocumented
def default_observers(update=None):

return a map from suite names to default observer names.

This function can also be used to update this map using a dict or a list of key-value pairs.

_default_observers: dict[str, str] =

Undocumented