for rn in range(5):
# rn == 0 no rotates
if rn == 0:
# output batch filenames
NAME_train = '{}_BATCH_{}_T{}_{}'.format(
var, size, domain, sea)
NAME_valid = '{}_BATCH_{}_V{}_{}'.format(
var, size, domain, sea)
# random cropping + batch gen
print('----- Training data process -----')
FEATURE_train = pu.random_cropping(input_train_3d,
keys_3d,
input_2d,
keys_2d,
mask[domain],
size,
gap,
ind_train,
rnd_range=2)
FEATURE_train = pu.feature_norm(FEATURE_train,
method=norm)
pu.batch_gen(FEATURE_train, batch_size, BATCH_dir,
NAME_train, 0)
print('----- Validation data process -----')
FEATURE_valid = pu.random_cropping(input_valid_3d,
keys_3d,
input_2d,
keys_2d,
mask[domain],
num_train[domain] = 0
num_valid[domain] = 0
# output batch filenames
NAME_train = '{}_BATCH_{}_T{}-{}_{}'.format(
var, size, source, domain, sea)
NAME_valid = '{}_BATCH_{}_V{}-{}_{}'.format(
var, size, source, domain, sea)
# random cropping + batch gen
# print('----- Training data process -----')
# FEATURE_train = pu.random_cropping(input_train_3d, keys_3d, input_2d, keys_2d, mask[domain],
# size, gap, ind_train, rnd_range=2, clim=False)
# FEATURE_train = pu.feature_norm(FEATURE_train, method=norm)
# pu.batch_gen(FEATURE_train, batch_size, BATCH_dir, NAME_train, 0);
print('----- Validation data process -----')
FEATURE_valid = pu.random_cropping(input_valid_3d,
keys_3d,
input_2d,
keys_2d,
mask[domain],
size,
gap,
ind_valid,
rnd_range=2,
clim=False)
FEATURE_valid = pu.feature_norm(FEATURE_valid, method=norm)
pu.batch_gen(FEATURE_valid, batch_size, BATCH_dir,
NAME_valid, 0)
num_train = {}; num_valid = {} # aug ind (numbering augmented batches)
for domain in domains:
num_train[domain] = 0
num_valid[domain] = 0
# batch augmentation and gen with 90-degree rotations
for rn in range(5):
# rn == 0 no rotates
if rn == 0:
# output batch filenames
NAME_train = '{}_BATCH_{}_T{}_{}'.format(var, size, domain, sea)
NAME_valid = '{}_BATCH_{}_V{}_{}'.format(var, size, domain, sea)
# random cropping + batch gen
print('----- Training data process -----')
FEATURE_train = pu.random_cropping(input_3d, keys_3d, input_2d, keys_2d, mask[domain], size, gap, ind_train, sparse_f=sparse_f)
FEATURE_train = pu.feature_norm(FEATURE_train, method=norm)
pu.batch_gen(FEATURE_train, batch_size, BATCH_dir, NAME_train, 0);
print('----- Validation data process -----')
FEATURE_valid = pu.random_cropping(input_3d, keys_3d, input_2d, keys_2d, mask[domain], size, gap, ind_valid, sparse_f=sparse_f)
FEATURE_valid = pu.feature_norm(FEATURE_valid, method=norm)
pu.batch_gen(FEATURE_valid, batch_size, BATCH_dir, NAME_valid, 0);
else:
print('Rotation round: {}'.format(rn))
# Feature rotations
for key in keys_3d:
input_3d[key] = np.rot90(input_3d[key], k=1, axes=(1, 2))
for key in keys_2d:
input_2d[key] = np.rot90(input_2d[key], k=1)