SEQ TINYIMAGENET#

Classes#

class datasets.seq_tinyimagenet.MyTinyImagenet(root, train=True, transform=None, target_transform=None, download=False)[source]#

Bases: TinyImagenet

Overrides the TinyImagenet dataset to change the getitem function.

class datasets.seq_tinyimagenet.SequentialTinyImagenet(args)[source]#

Bases: ContinualDataset

The Sequential Tiny Imagenet dataset.

Parameters:

  • NAME (str) – name of the dataset.

  • SETTING (str) – setting of the dataset.

  • N_CLASSES_PER_TASK (int) – number of classes per task.

  • N_TASKS (int) – number of tasks.

  • N_CLASSES (int) – number of classes.

  • SIZE (tuple) – size of the images.

  • MEAN (tuple) – mean of the dataset.

  • STD (tuple) – standard deviation of the dataset.

  • TRANSFORM (torchvision.transforms) – transformations to apply to the dataset.

MEAN = (0.4802, 0.448, 0.3975)#
NAME: str = 'seq-tinyimg'#
N_CLASSES: int = 200#
N_CLASSES_PER_TASK: int = 20#
N_TASKS: int = 10#
SETTING: str = 'class-il'#
SIZE: Tuple[int] = (64, 64)#
STD = (0.277, 0.2691, 0.2821)#
TRANSFORM = Compose(     RandomCrop(size=(64, 64), padding=4)     RandomHorizontalFlip(p=0.5)     ToTensor()     Normalize(mean=(0.4802, 0.448, 0.3975), std=(0.277, 0.2691, 0.2821)) )#
get_backbone()[source]#
get_batch_size()[source]#
get_class_names()[source]#
get_data_loaders()[source]#
Return type:

Tuple[DataLoader, DataLoader]

static get_denormalization_transform()[source]#
get_epochs()[source]#
static get_loss()[source]#
static get_normalization_transform()[source]#
get_transform()[source]#
class datasets.seq_tinyimagenet.TinyImagenet(root, train=True, transform=None, target_transform=None, download=False)[source]#

Bases: Dataset

Defines the Tiny Imagenet dataset.