def define_criterion(self, name):
if name.lower() == 'bce+dice':
self.criterion = Loss.BCE_Dice()
elif name.lower() == 'dice':
self.criterion = Loss.DiceLoss()
elif name.lower() == 'bce':
self.criterion = nn.BCEWithLogitsLoss()
elif name.lower() == 'robustfocal':
self.criterion = Loss.RobustFocalLoss2d()
elif name.lower() == 'lovasz-hinge' or name.lower() == 'lovasz':
self.criterion = Loss.Lovasz_Hinge(per_image=True)
elif name.lower() == 'bce+lovasz':
self.criterion = Loss.BCE_Lovasz(per_image=True)
else:
raise NotImplementedError(
'Loss {} is not implemented'.format(name))
def define_criterion(self, name):
if name.lower() == 'bce+dice':
self.criterion = Loss.BCE_Dice()
elif name.lower() == 'dice':
self.criterion = Loss.DiceLoss()
elif name.lower() == 'bce':
# self.criterion = nn.BCEWithLogitsLoss()
# self.criterion = nn.CrossEntropyLoss(size_average=False, reduction='sum')
self.criterion = nn.CrossEntropyLoss()
# self.criterion = Loss.CE_SOFT()
elif name.lower() == 'robustfocal':
self.criterion = Loss.RobustFocalLoss2d()
elif name.lower() == 'lovasz-hinge' or name.lower() == 'lovasz':
self.criterion = Loss.Lovasz_Hinge(per_image=True)
elif name.lower() == 'bce+lovasz':
self.criterion = Loss.BCE_Lovasz(per_image=True)
else:
raise NotImplementedError(
'Loss {} is not implemented'.format(name))
def __init__(self, config, train_loader, valid_loader, test_loader):
# Data loader
self.train_loader = train_loader
self.valid_loader = valid_loader
self.test_loader = test_loader
# Models
self.unet = None
self.optimizer = None
self.scheduler = None
self.img_ch = config.img_ch
self.output_ch = config.output_ch
# Losses
self.criterion = Loss.DiceLoss()
# Hyper-parameters
self.lr = config.lr
self.beta1 = config.beta1
self.beta2 = config.beta2
# Training settings
self.num_epochs = config.num_epochs
self.num_epochs_test = config.num_epochs_test
self.batch_size = config.batch_size
# Path
self.model_path = config.model_path
self.train_result_path = config.train_result_path
self.val_result_path = config.val_result_path
self.test_result_path = config.test_result_path
self.mode = config.mode
self.device = torch.device(
'cuda' if torch.cuda.is_available() else 'cpu')
self.model_type = config.model_type
self.build_model()
