def path_message(self):
"""Path options values"""
def add_message(part1, part2):
left = " " * (round(TOTAL_LENGTH / 3) - len(part2 + ':'))
right = getattr(part1, part2) + '\n'
if TOTAL_LENGTH >= (len(left + part2 + ': ' + right) + 6):
left = ">>>" + " " * (round(TOTAL_LENGTH / 3) -
len(part2 + ':') - 3)
right = getattr(part1, part2) + " " * (round(
TOTAL_LENGTH - 3 - len(left + part2 + ': ' +
getattr(part1, part2)))) + "<<<\n"
return left + part2 + ': ' + right
message = ''
equal_left, equal_right = cal_equal(len(self.mode) + 8)
message += '\n' + ('=' * equal_left) + ' ' + self.mode + ' Paths ' + (
'=' * equal_right) + '\n'
for a in dir(self):
if self.mode == 'Train':
if not a.startswith("__") and not callable(getattr(self, a)) and \
a.split('_')[-1] == 'DIR' and 'RESULT' not in a and 'TEST' not in a:
message += add_message(self, a)
elif self.mode == 'Test':
if not a.startswith("__") and not callable(getattr(self, a)) and \
a.split('_')[-1] == 'DIR' and 'TRAIN' not in a and 'VAL' not in a:
message += add_message(self, a)
equal_left, equal_right = cal_equal(5)
message += '=' * equal_left + ' End ' + '=' * equal_right
return message
def print_test_info(cfg, step, metrics):
"""Print validate information on the screen.
Inputs:
cfg: training options
step: a list includes --> [per_step, total_step]
metrics: the current batch testing metrics
"""
report = info = message = ""
if step[0] == 0:
equal_left, equal_right = cal_equal(15)
info += "\n" + "=" * equal_left + " Start Testing " + "=" * equal_right
print(info)
if cfg.opts.save_test_log:
wrote_txt_file(os.path.join(cfg.CHECKPOINT_DIR, 'Test_log.txt'),
info,
mode='w',
show=False)
progress_bar(step[0], step[1], display=False)
if cfg.opts.test_label != 'None':
if step[0] + 1 >= step[1]:
report, message = prepare_metrics(report, message, metrics)
if cfg.opts.save_test_log:
wrote_txt_file(os.path.join(cfg.CHECKPOINT_DIR,
'Test_log.txt'),
message + "\n" + report,
mode='a',
show=False)
def print_train_info(val_flag, cfg, epoch, step, loss, lr, metrics):
"""Print training epoch, step, loss, lr, and auc information on the screen.
Inputs:
val_flag: whether print the val info according to the epoch(bool)
cfg: training options
epoch: a list includes --> [start_epoch, per_epoch, total_epoch]
step: a list includes --> [per_step, total_step]
loss: a float includes --> train loss value
lr: the current learning rate
metrics: the current batch training metrics
"""
message = ""
# print '==== Start training ===='
if epoch[1] == 1 and step[0] == 1:
equal_left, equal_right = cal_equal(16)
message += "\n" + "=" * equal_left + " Start Training " + "=" * equal_right
# print '==== Start training ===='
elif epoch[1] == epoch[0] and epoch[1] != 1 and step[0] == 1:
equal_left, equal_right = cal_equal(19)
message += "\n" + "=" * equal_left + " Continue Training " + "=" * equal_right
# print '---- Epoch [1/10] ----'
if (epoch[1] % cfg.opts.print_epoch == 0 or epoch[1] == epoch[0]
or epoch[1] == epoch[2]) and step[0] - 1 == 0:
val_flag = True
info = " Epoch [{}/{}] ".format(epoch[1], epoch[2])
equal_left, equal_right = cal_equal(len(info))
message += "\n" + "-" * equal_left + info + "-" * equal_right
message += "\n>>> Learning rate {:.7f}\n".format(lr)
# print time, per_step, loss and acc
if val_flag and (step[0] - 1 == 0 or step[0] == step[1]
or step[0] % cfg.opts.print_step == 0):
current_time = time.strftime("%m-%d %H:%M:%S", time.localtime())
message += "{} Step:[{}/{}]".format(current_time, step[0], step[1])
message += " " * (len(str(step[1])) -
len(str(step[0]))) + " Loss:{:.4f} ".format(loss)
for metric in metrics:
message += "{}:{:.3f}% ".format(metric, metrics[metric] * 100)
if len(message) != 0:
print(message)
if cfg.opts.save_train_log:
mode = 'w' if epoch[1] == 1 and step[0] == 1 else 'a'
wrote_txt_file(os.path.join(cfg.CHECKPOINT_DIR, 'Train_Log.txt'),
message,
mode=mode,
show=False)
return val_flag
def param_message(self, args):
"""Parameter options values."""
message = ''
equal_left, equal_right = cal_equal(len(self.mode) + 10)
message += '\n' + ('=' *
equal_left) + ' ' + self.mode + ' Options ' + (
'=' * equal_right) + '\n'
for k, v in sorted(vars(args).items()):
if 'checkpoint' in k:
continue
left = " " * (round(TOTAL_LENGTH / 2) - len(str(k) + " :"))
right = " " + str(v) + "\n"
if TOTAL_LENGTH >= (len(left + str(k) + " :" + right) + 6):
left = ">>>" + " " * (round(TOTAL_LENGTH / 2) -
len(str(k) + " :") - 3)
right = " " + str(v) + " " * (round(TOTAL_LENGTH - 3 - len(
left + str(k) + " : " + str(v)))) + "<<<\n"
message += left + str(k) + " :" + right
equal_left, equal_right = cal_equal(5)
message += '=' * equal_left + ' End ' + '=' * equal_right
return message
def _display_network(self, verbose=False):
"""Print the total number of parameters and architecture
in the network.
Inputs:
verbose: print the network architecture or not(bool)
"""
equal_left, equal_right = cal_equal(22)
print("\n" + "=" * equal_left + " Networks Initialized " +
"=" * equal_right)
num_params = 0
for param in self.network.parameters():
num_params += param.numel()
if verbose:
print(self.network)
print('>>> [%s] Total size of parameters : %.3f M' %
(self.opts.net_name, num_params / 1e6))
print('>>> [%s] Weights initialize with : %s ' %
(self.opts.net_name, self.opts.init_type))
if self.cfg.mode == 'Train':
print('>>> [%s] Learning rate scheduler : %s ' %
(self.opts.net_name, self.opts.lr_scheduler))
equal_left, equal_right = cal_equal(6)
print('%s Done %s' % ('=' * equal_left, '=' * equal_right))
print('>>> [%s] was created ...' % type(self).__name__)
def _display_network(self, verbose=False):
"""Print the total number of parameters and architecture
in the network.
Inputs:
verbose: print the network architecture or not(bool)
"""
equal_left, equal_right = cal_equal(22)
print("\n" + "=" * equal_left + " Networks Initialized " +
"=" * equal_right)
if verbose:
self.network.summary()
if self.cfg.mode == 'Train':
print('>>> [%s] Learning rate scheduler : %s ' %
(self.opts.net_name, self.opts.lr_scheduler))
print('>>> [%s] was created ...' % type(self).__name__)
def load_data(self, mode, shuffle=False):
"""Load the train or val or test dataset"""
if mode == 'Train':
label_name, label_dir, data_dir = [
self.opts.train_label, self.cfg.TRAIN_LABEL_DIR,
self.cfg.TRAIN_DATA_DIR
]
elif mode == 'Val':
label_name, label_dir, data_dir = [
self.opts.val_label, self.cfg.VAL_LABEL_DIR,
self.cfg.VAL_DATA_DIR
]
else:
label_name, label_dir, data_dir = [
self.opts.test_label, self.cfg.TEST_LABEL_DIR,
self.cfg.TEST_DATA_DIR
]
if label_name == 'None':
label_data = []
test_names = os.listdir(data_dir)
for name in test_names:
label_data.append([name, 0])
else:
label_data = self._open_data_file(label_name, label_dir)
if shuffle:
random.shuffle(label_data)
for index, data_set in enumerate(label_data):
if mode == 'Test':
length = self.opts.num_test if self.opts.num_test < len(
label_data) else len(label_data)
if index + 1 > self.opts.num_test:
break
else:
length = len(label_data)
progress_bar(index, length, "Loading {} dataset".format(mode))
self._add_to_database(index, data_set, data_dir)
equal_left, equal_right = cal_equal(6)
print('\n%s Done %s' % ('=' * equal_left, '=' * equal_right))
def print_test_info(cfg, step, loss=100.0, metric=0.0):
"""Print validate information on the screen.
Inputs:
cfg: training options
step: a list includes --> [per_step, total_step]
loss: a float includes --> val loss value
metric: the current batch testing metric
"""
info = message = ""
if step[0] == 0:
equal_left, equal_right = cal_equal(15)
info += "\n" + "=" * equal_left + " Start Testing " + "=" * equal_right
print(info)
if cfg.opts.save_test_log:
wrote_txt_file(os.path.join(cfg.CHECKPOINT_DIR, 'Test_log.txt'), info, mode='w', show=False)
progress_bar(step[0], step[1], display=False)
if cfg.opts.test_label != 'None':
if step[0] + 1 >= step[1]:
message += "\n>>> Loss:{:.4f} ACC:{:.3f}% ".format(loss / step[1], metric / step[1] * 100)
print(message)
if cfg.opts.save_test_log:
wrote_txt_file(os.path.join(cfg.CHECKPOINT_DIR, 'Test_log.txt'), message, mode='a', show=False)