Getting started with COCO

The COCO framework is mainly divided into two main parts which can be used independently.

Getting started: running an algorithm on a COCO benchmark suite

• In Python: see Getting started—Benchmarking an algorithm (collecting data) with Python.
• In other languages: see the instructions in code-experiments/build and in the respective subfolder. This usually requires to make a local clone of github/numbbo/coco and run scripts/fabricate, see here.

Getting started: displaying COCO results (postprocess)

See Getting started–Analysing and comparing results (postprocess).