def run_genetic_algorithm(params_tuple):
"""Main function to run genetic algorithm with prefix number.
params_tuple : tuple of 3 objects:
number : number of GA to run (moreover its prefix).
log_file : log file to write output.
shared_dict : dictionary to share information between processes.
"""
np.random.seed()
number = 'ID'
try:
def write_func(string): return support.write_log(log_file, string,
write_to_stdout=not params.silence)
number, params, log_file, shared_dict = params_tuple
write_func('Run genetic algorithm number ' + str(number))
ga_instance = GA(params, prefix=str(number))
ga_instance.run(shared_dict=shared_dict)
write_func('Finish genetic algorithm number ' + str(number))
return number, ga_instance.best_model()
except Exception, e:
err_str = 'GA number ' + str(number) + ': ' + str(e)
support.error(err_str, error_instance=e, exit=False)
raise RuntimeError(err_str)
def load_parameters_from_python_file(filename, as_module=False):
if not as_module:
f = support.check_file_existence(filename)
try:
module = imp.load_source('module', f)
except Exceprint, e:
support.error('File ' + filename + " is not valid python file.",
error_instance=e)
def main():
parser = argparse.ArgumentParser(
'GADMA module for runs of local search on bootstrapped data. Is needed for calculating confidence intervals.'
)
parser.add_argument('-b',
'--boots',
metavar="<dir>",
required=True,
help='Directory where bootstrapped data is located.')
parser.add_argument(
'-d',
'--dem_model',
metavar="<filename>",
required=True,
help=
'File with demographic model. Should contain `model_func` or `generated_model` function. One can put there several extra parameters and they will be taken automatically, otherwise one will need to enter them manually. Such parameters are:\n\t1) p0 (or popt) - initial parameters values\n\t2) lower_bound - list of lower bounds for parameters values\n\t4) upper_bound - list of upper bounds for parameters values\n\t5) par_labels/param_labels - list of string names for parameters 6) pts - pts for dadi (if there is no pts then moments will be run automatically).'
)
parser.add_argument('-o',
'--output',
metavar="<dir>",
required=True,
help='Output directory.')
parser.add_argument('-j',
'--jobs',
metavar="N",
type=int,
default=1,
help='Number of threads for parallel run.')
parser.add_argument(
'--opt',
metavar="log/powell",
type=str,
default='log',
help=
'Local search algorithm, by now it can be:\n\t1) `log` - Inference.optimize_log\n\t2) `powell` - Inference.optimize_powell.'
)
parser.add_argument(
'-p',
'--params',
metavar="<filename>",
type=str,
default=None,
help=
'Filename with parameters, should be valid python file. Parameters are presented in -d/--dem_model option description upper.'
)
args = parser.parse_args()
output = support.ensure_dir_existence(args.output, False)
all_boot = gadma.Inference.load_bootstrap_data_from_dir(
args.boots, return_filenames=True)
print(str(len(all_boot)) + ' bootstrapped data found.')
#extract model_func, we cannot put it to function as multiprocessing need pickleable functions
args.dem_model = support.check_file_existence(args.dem_model)
try:
file_with_model_func = imp.load_source('module', args.dem_model)
except Exception, e:
support.error('File ' + args.dem_model + " is not valid python file.",
error_instance=e)
args.dem_model = support.check_file_existence(args.dem_model)
try:
file_with_model_func = imp.load_source('module', args.dem_model)
except Exception, e:
support.error('File ' + args.dem_model + " is not valid python file.",
error_instance=e)
try:
model_func = file_with_model_func.model_func
name = 'model_func'
except:
try:
model_func = file_with_model_func.generated_model
name = 'generated_model'
except:
support.error(
"File " + dem_model_filename +
' does not contain function named `model_func` or `generated_model`.'
)
par_labels = get_params_names(args.dem_model, name)
lower_bound, upper_bound, p0, new_par_labels, pts = load_parameters_from_python_file(
file_with_model_func, as_module=True)
if new_par_labels is not None:
par_labels = new_par_labels
if args.params is not None:
new_lower_bound, new_upper_bound, new_p0, new_par_labels, new_pts = load_parameters_from_python_file(
args.params)
if new_lower_bound is not None:
lower_bound = new_lower_bound
if new_upper_bound is not None:
upper_bound = new_upper_bound
if new_p0 is not None:
def main():
params = options.parse_args()
log_file = os.path.join(
params.output_dir, 'GADMA.log')
open(log_file, 'w').close()
support.write_log(log_file, "--Successful arguments' parsing--\n")
params_filepath = os.path.join(
params.output_dir, 'params')
params.save(params.output_dir)
if not params.test:
support.write_log(
log_file, 'You can find all parameters of this run in:\t\t' + params_filepath + '\n')
support.write_log(log_file, 'All output is saved (without warnings and errors) in:\t' +
os.path.join(params.output_dir, 'GADMA.log\n'))
support.write_log(log_file, '--Start pipeline--\n')
# For debug
# run_genetic_algorithm((1, params, log_file, None))
# Create shared dictionary
m = Manager()
shared_dict = m.dict()
# Start pool of processes
start_time = datetime.now()
pool = Pool(processes=params.processes,
initializer=worker_init)
try:
pool_map = pool.map_async(
run_genetic_algorithm,
[(i + 1, params, log_file, shared_dict)
for i in range(params.repeats)])
pool.close()
precision = 1 - int(math.log(params.epsilon, 10))
# graceful way to interrupt all processes by Ctrl+C
min_counter = 0
while True:
try:
multiple_results = pool_map.get(
60 * params.time_for_print)
break
# catch TimeoutError and get again
except multiprocessing.TimeoutError as ex:
print_best_solution_now(
start_time, shared_dict, params, log_file,
precision, draw_model=params.matplotlib_available)
except Exception, e:
pool.terminate()
support.error(str(e))
print_best_solution_now(start_time, shared_dict, params,log_file,
precision, draw_model=params.matplotlib_available)
support.write_log(log_file, '\n--Finish pipeline--\n')
if params.test:
support.write_log(log_file, '--Test passed correctly--')
if params.theta is None:
support.write_log(
log_file, "\nYou didn't specify theta at the beginning. If you want change it and rescale parameters, please see tutorial.\n")
if params.resume_dir is not None and (params.initial_structure != params.final_structure).any():
support.write_log(
log_file, '\nYou have resumed from another launch. Please, check best AIC model, as information about it was lost.\n')
support.write_log(log_file, 'Thank you for using GADMA!')
break
# catch TimeoutError and get again
except multiprocessing.TimeoutError as ex:
print_best_solution_now(
start_time, shared_dict, params, log_file,
precision, draw_model=params.matplotlib_available)
except Exception, e:
pool.terminate()
support.error(str(e))
print_best_solution_now(start_time, shared_dict, params,log_file,
precision, draw_model=params.matplotlib_available)
support.write_log(log_file, '\n--Finish pipeline--\n')
if params.test:
support.write_log(log_file, '--Test passed correctly--')
if params.theta is None:
support.write_log(
log_file, "\nYou didn't specify theta at the beginning. If you want change it and rescale parameters, please see tutorial.\n")
if params.resume_dir is not None and (params.initial_structure != params.final_structure).any():
support.write_log(
log_file, '\nYou have resumed from another launch. Please, check best AIC model, as information about it was lost.\n')
support.write_log(log_file, 'Thank you for using GADMA!')
except KeyboardInterrupt:
pool.terminate()
support.error('KeyboardInterrupt')
if __name__ == '__main__':
main()
