while True:
seed = randint(0, 2**32-1)
set_random_seed(seed)
set_numpy_seed(seed)
for i_param, param in enumerate(parameters):
last_perfs = 0
min_perfs = 0
time = 0
for i_cv, cv in enumerate(cvs):
if i_param + i_cv == 0:
model_storage = 'save'
else:
model_storage = 'load'
last_perf, min_perf, dt = Tr.train_network(param, cv, seed=seed, callbacks=cbs, verbose=False, model_path=model_path, model_storage=model_storage)
last_perfs += last_perf
min_perfs += min_perf
time += dt
last_perfs /= n_cv
min_perfs /= n_cv
row = [i_param, seed]
row += pg.as_array(param)
row += [str(c) for c in [last_perfs, min_perfs, time]]
print(row)
log.write_row(row)
pass
head += ['last_perf', 'min_perf', 'time']
print(head)
log = Bu.CSVWriter(filename, head=head)
shuffle(parameters)
for i_param, param in enumerate(parameters):
last_perfs = 0
min_perfs = 0
time = 0
for i_cv, cv in enumerate(cvs):
if param['optimizer'] == 'adadelta':
last_perf, min_perf, dt = Tr.train_network(param,
cv,
seed=seed,
callbacks=cbs[1:],
verbose=False)
else:
last_perf, min_perf, dt = Tr.train_network(param,
cv,
seed=seed,
callbacks=cbs,
verbose=False)
last_perfs += last_perf
min_perfs += min_perf
time += dt
last_perfs /= n_cv
min_perfs /= n_cv
row = [i_param]