Optimize Parameters GA

benchmark_ea.python.optimize_parameters_genetic_alg.create_cpu_optimizer(args, logger)

returns configured bluepyopt.optimisations.DEAPOptimisation

benchmark_ea.python.optimize_parameters_genetic_alg.get_parser()

Get parsed arguemnts from command line. Following arguments are :param continue: (bool) should EA continue from checkpoint :param checkpoint: (str) path to BluePyOpt formulated checkpoint :param offspring_size: (int) number of indivduals to use in EA :param max_ngen: (int) number of generations to run to complete EA :param n_stims: (int) number of stimuli to use in EA :param n_sfs: (int) number of score functions to use in EA :param ipyparallel: (bool) use ipyParallel mapping function

benchmark_ea.python.optimize_parameters_genetic_alg.main(pool)

Run optimization for NeuroGPU-EA :param pool: (multiprocessing.pool) pool of cpu process is created before calling main. This was necessary in ppc64le env. on summit.

benchmark_ea.python.optimize_parameters_genetic_alg.my_record_stats(stats, logbook, gen, population, invalid_count)

Update the statistics with the new population :param logbook: (deap.tools.logbook) DEAP logbook

benchmark_ea.python.optimize_parameters_genetic_alg.my_update(halloffame, history, population)

Custom update to bluepyopt / DEAP that allows us to save intermediate results by calling save_logs

benchmark_ea.python.optimize_parameters_genetic_alg.save_logs(fn, best_indvs, population)

Save best individuals as a pickle file in best_indv_logs