Hoc-Evaluators

class benchmark_ea.python.hoc_evaluator_neuron.hoc_evaluator(args, logger=None)

evaluate_with_lists(param_values)

Evaluate parameter a parameter set (abstract).

Args:

params (list of Parameters):

The parameter values to be evaluated.

Returns:

objectives (list of Objectives):

List of Objectives with values calculated by the Evaluator.

benchmark_ea.python.hoc_evaluator_neuron.top_SFs(max_sfs=0, max_stims=6)

finds scoring functions w/ weight over 50 and pairs them with that stim and sends them to mapping function so that we will run so many processes Arguments ————————————————————– run_num: the number of times neuroGPU has ran for 8 stims, keep track of what stims we are picking out score functions for

class benchmark_ea.python.hoc_evaluator_neurogpu.hoc_evaluator(pool, n_stims, n_sfs, sf_module='efel', logger=None)

evaluate_with_lists(param_values)

This function overrides the BPOP built in function. It is currently set up to run GPU tasks for each stim in chunks based on number of GPU resources then stacks these results and sends them off to be evaluated. It runs concurrently so that while nGpus are busy, results ready for evaluation are evaluated. Parameters ——————– param_values: Population sized list of parameter sets to be ran through neruoGPU then scored and evaluated

2d list of scalar scores for each parameter set w/ shape (nindv,1)

getVolts(idx)

Helper function that gets volts from data and shapes them for a given stim index

map_par(run_num)

This function maps out what stim and score function pairs should be mapped to be evaluated in parallel first it finds the pairs with the highest weights, the maps them and then adds up the score for each stim for every individual.

run_num: the amount of times neuroGPU has ran for 8 stims

2d list of scalar scores for each parameter set w/ shape (nindv,nstims)

my_evaluate_invalid_fitness(population)

Evaluate the individuals with an invalid fitness Returns the count of individuals with invalid fitness

old_top_SFs(run_num)

finds scoring functions w/ weight over 50 and pairs them with that stim and sends them to mapping function so that we will run so many processes Arguments ————————————————————– run_num: the number of times neuroGPU has ran for 8 stims, keep track of what stims we are picking out score functions for

run_model(stim_ind, params)

stim_ind: index to send as arg to neuroGPU params: DEPRECATED remove

p_object: process object that stops when neuroGPU done

top_SFs(run_num, max_sfs=0)

finds scoring functions w/ weight over 50 and pairs them with that stim and sends them to mapping function so that we will run so many processes Arguments ————————————————————– run_num: the number of times neuroGPU has ran for 8 stims, keep track of what stims we are picking out score functions for

class benchmark_ea.python.hoc_evaluator_coreneuron.hoc_evaluator(pool, n_stims, n_sfs, sf_module='efel', logger=None)

CoreNeuron hoc evaluator class Args

param pool

(multiprocessing.Pool) pool of cpu processers to use for multiprocessing. This parameter is deprecated on Cori (x86_64) but not on ppc64le.

param n_stims

(int) number of stimuli to use in EA

param n_sfs

(int) number of score functions to use in EA

param sf_module

(optional str) score function module to use in EA.

evaluate_with_lists(param_values)

This function overrides the BPOP built in function. It is currently set up to run GPU tasks for each stim in chunks based on number of GPU resources then stacks these results and sends them off to be evaluated. It runs concurrently so that while nGpus are busy, results ready for evaluation are evaluated. Parameters ——————– param_values: Population sized list of parameter sets to be ran through neruoGPU then scored and evaluated

2d list of scalar scores for each parameter set w/ shape (nindv,1)

map_par(run_num)

This function maps out what stim and score function pairs should be mapped to be evaluated in parallel first it finds the pairs with the highest weights, the maps them and then adds up the score for each stim for every individual.

run_num: the amount of times neuroGPU has ran for 8 stims

2d list of scalar scores for each parameter set w/ shape (nindv,nstims)

my_evaluate_invalid_fitness(population)

Evaluate the individuals with an invalid fitness Returns the count of individuals with invalid fitness

benchmark_ea.python.hoc_evaluator_coreneuron.run_model(pc)

Function to run h.finitialize and psolve using parallel context from Coreneuron