cocopp.algportfolio
module documentationcocopp
Algorithm portfolio data set module. The algorithm portfolio consists in running multiple algorithms in parallel. Current limitation: the portfolio data set must come from data sets that are identical (same number of repetitions on the same instances of the functions. **Example:** .. plot:: :width: 75% import glob from pylab import * import pickle import cocopp import cocopp.compall.pprldmany import cocopp.algportfolio # Collect and unarchive data dsets = {} for alg in ('BIPOP-CMA-ES_hansen_noiseless', 'NEWUOA_ros_noiseless'): dsets[alg] = cocopp.load(cocopp.bbob(alg)) # Generate the algorithm portfolio dspf = cocopp.algportfolio.build(dsets) dsets['Portfolio'] = dspf # store the portfolio in dsets # plot the run lengths distribution functions plt.figure() for algname, ds in dsets.items(): dataset = ds.dictByDimFunc()[10][13] # DataSet dimension 10 on F13 cocopp.compall.pprldmany.plot(dataset, label=algname) cocopp.compall.pprldmany.beautify() legend(loc='best') # Display legend plt.show()
Class | Usage | Undocumented |
Class | DataSet | Unit element of algorithm portfolio data set. |
Function | build | Merge datasets in an algorithm portfolio. |
Merge datasets in an algorithm portfolio. :param dict dictAlg: dictionary of data sets with algorithm name for keys, see ``pproc.DataSetList.dictByAlg`` :param seq sortedAlgs: sequence for sorting the entries of :py:data:`dictAlg`, if not provided, dictAlg.keys() will be instead :returns: an instance of :py:class:`DataSetList` with the porfolio data sets