def create_gens(train, test):
gen, test_gen = None, None
if train != None:
gen = IQ_Generator(data_points=train,
dim=input_dims,
batch_size=bs,
n_classes=1,
shuffle=True,
augment=True,
distort=False,
dist_scale=.05,
flip=False,
permute=False,
gauss_noise=.001,
rand_seg_remove=3)
if test != None:
test_gen = IQ_Generator(data_points=test,
dim=input_dims,
batch_size=1,
n_classes=1,
shuffle=False,
augment=False)
if gen != None and test_gen != None:
return gen, test_gen
elif gen != None:
return gen
elif test_gen != None:
return test_gen
def create_test_gen(test):
test_gen = IQ_Generator(data_points=test,
dim=input_dims,
batch_size=1,
n_classes=1,
shuffle=False,
augment=False)
return test_gen
def create_gens(train, test):
gen = IQ_Generator(data_points = train,
dim=input_dims,
batch_size = bs,
n_classes = 1,
shuffle = True,
augment = False,
distort = False,
dist_scale = .15,
permute = False,
gauss_noise = .01
)
test_gen = IQ_Generator(data_points = test,
dim=input_dims,
batch_size = bs,
n_classes = 1,
shuffle = False,
augment = False)
return gen, test_gen
def create_train_gen(train):
gen = IQ_Generator(data_points=train,
dim=input_dims,
batch_size=bs,
n_classes=1,
shuffle=False,
augment=False,
distort=True,
dist_scale=.05,
gauss_noise=.001,
rand_seg_remove=0)
return gen