kernel_size=kernel_size,
filter_size=filter_size,
max_pool=max_pool,
dense_size=dense_size,
conv_type=conv_type,
batch_norm=batch_norm,
kernel_regularizer=kernel_regularizer)
model.summary()
key = input('Continue [Y/n]')
if key == 'n':
break
#Create IDs
model_id = "{:s}{:s}{}".format(
str(lr).replace('.', 'e'), optimizer, int(time()))
dataset_id = ids.get_dataset_id(dataset_dir)
print("-" * 30)
print("{:s}".format(dataset_id))
print("{:s}".format(model_id))
print("-" * 30)
final_id = data_type + scan_type + model_id + '_' + net_id
#Write Summary--------------------------------------------
# Open the file
cwd = os.getcwd() #directory where the script is called by terminal
with open('{}/summary{}.txt'.format(cwd, final_id), 'w') as fh:
# Pass the file handle in as a lambda function to make it callable
model.summary(print_fn=lambda x: fh.write(x + '\n'))
model.compile(loss='binary_crossentropy',
optimizer=opti,
pfa_re = re.compile(r"P\de\d+")
dataset_id_re = re.compile(r"Tex\d+\_\d+")
data_re = re.compile(r"[a-zA-Z]+$")
fail_re = re.compile(r"aaandnadnadnan")
list_re = []
list_re.append(num_acq_re)
list_re.append(pfa_re)
list_re.append(dataset_id_re)
list_re.append(data_re)
list_re.append(data_re)
list_re.append(fail_re)
#for rr in list_re:
# print (rr.search(dataset_dir).group())
print(ids.get_dataset_id(dataset_dir))
#%%
str_in = 'B_FA566_2017_07_07_003Swath1x0_ch0_y0.png'
str_in = 'B_T66_2017_07_07_003Swath1x0_ch0_y0.png'
import re
anomaly_id = re.compile(r"\_(T|FA)\d+\_")
acq_id = re.compile(r"_\d\d\d\d\_\d\d_\d\d[^x]*")
is_central = re.compile(r"x0\_ch0\_y0\.")
list_re = []
list_re.append(anomaly_id)
list_re.append(acq_id)
list_re.append(is_central)
for rr in list_re: