def print_loss_log(self, start_time, iters_per_epoch, e, i, loss):
"""
Prints the loss and elapsed time for each epoch
"""
total_iter = self.num_epochs * iters_per_epoch
cur_iter = e * iters_per_epoch + i
elapsed = time.time() - start_time
total_time = (total_iter - cur_iter) * elapsed / (cur_iter + 1)
epoch_time = (iters_per_epoch - i) * elapsed / (cur_iter + 1)
epoch_time = str(datetime.timedelta(seconds=epoch_time))
total_time = str(datetime.timedelta(seconds=total_time))
elapsed = str(datetime.timedelta(seconds=elapsed))
log = "Elapsed {}/{} -- {}, Epoch [{}/{}], Iter [{}/{}], " \
"loss: {:.4f}".format(elapsed,
epoch_time,
total_time,
e + 1,
self.num_epochs,
i + 1,
iters_per_epoch,
loss)
write_print(self.output_txt, log)
def test(self):
"""
Evaluates the performance of the model using the test dataset
"""
top_1_correct, top_5_correct, total = self.eval(self.data_loader)
log = "top_1_acc: {:.4f}--top_5_acc: {:.4f}".format(
top_1_correct / total, top_5_correct / total)
write_print(self.output_txt, log)
def load_pretrained_model(self):
"""
loads a pre-trained model from a .pth file
"""
self.model.load_state_dict(
torch.load(
os.path.join(self.model_save_path,
'{}.pth'.format(self.pretrained_model))))
write_print(self.output_txt,
'loaded trained model {}'.format(self.pretrained_model))
def train_evaluate(self, e):
"""
Evaluates the performance of the model using the train dataset
"""
top_1_correct, top_5_correct, total = self.eval(self.data_loader)
log = "Epoch [{}/{}]--top_1_acc: {:.4f}--top_5_acc: {:.4f}".format(
e + 1, self.num_epochs, top_1_correct / total,
top_5_correct / total)
write_print(self.output_txt, log)
return top_1_correct / total, top_5_correct / total
def print_network(self, model, name):
"""
Prints the structure of the network and the total number of parameters
"""
num_params = 0
for p in model.parameters():
num_params += p.numel()
write_print(self.output_txt, name)
write_print(self.output_txt, str(model))
write_print(self.output_txt,
'The number of parameters: {}'.format(num_params))
# dataset
# parser.add_argument('--data_path', type=str, default='../data/c256/')
# parser.add_argument('--train_data_path', type=str,
# default='caltech_256_60_train_nobg_norm.hdf5')
# parser.add_argument('--train_x_key', type=str, default='train_x')
# parser.add_argument('--train_y_key', type=str, default='train_y')
# parser.add_argument('--test_data_path', type=str,
# default='caltech_256_60_test_nobg_norm.hdf5')
# parser.add_argument('--test_x_key', type=str, default='test_x')
# parser.add_argument('--test_y_key', type=str, default='test_y')
# path
parser.add_argument('--model_save_path', type=str, default='./weights',
help='Path for saving weights')
# epoch step size
parser.add_argument('--loss_log_step', type=int, default=1)
parser.add_argument('--model_save_step', type=int, default=1)
parser.add_argument('--train_eval_step', type=int, default=1)
config = parser.parse_args()
args = vars(config)
print(args)
write_print('hello.txt', '------------ Options -------------')
for k, v in args.items():
write_print('hello.txt', '{}: {}'.format(str(k), str(v)))
write_print('hello.txt', '-------------- End ----------------')
# main(version, config)
def train(self):
"""
Training process
"""
self.losses = []
self.top_1_acc = []
self.top_5_acc = []
iters_per_epoch = len(self.data_loader)
# start with a trained model if exists
if self.pretrained_model:
start = int(self.pretrained_model.split('/')[-1])
else:
start = 0
# start training
start_time = time.time()
for e in range(start, self.num_epochs):
for i, (images, labels) in enumerate(tqdm(self.data_loader)):
images = to_var(images, self.use_gpu)
labels = to_var(torch.LongTensor(labels), self.use_gpu)
loss = self.model_step(images, labels)
# print out loss log
if (e + 1) % self.loss_log_step == 0:
self.print_loss_log(start_time, iters_per_epoch, e, i, loss)
self.losses.append((e, loss))
# save model
if (e + 1) % self.model_save_step == 0:
self.save_model(e)
# evaluate on train dataset
# if (e + 1) % self.train_eval_step == 0:
# top_1_acc, top_5_acc = self.train_evaluate(e)
# self.top_1_acc.append((e, top_1_acc))
# self.top_5_acc.append((e, top_5_acc))
# print losses
write_print(self.output_txt, '\n--Losses--')
for e, loss in self.losses:
write_print(self.output_txt, str(e) + ' {:.4f}'.format(loss))
# print top_1_acc
write_print(self.output_txt, '\n--Top 1 accuracy--')
for e, acc in self.top_1_acc:
write_print(self.output_txt, str(e) + ' {:.4f}'.format(acc))
# print top_5_acc
write_print(self.output_txt, '\n--Top 5 accuracy--')
for e, acc in self.top_5_acc:
write_print(self.output_txt, str(e) + ' {:.4f}'.format(acc))