SEQ GTRSRB

Classes

class datasets.seq_gtrsrb.MyGTSRB(root, split='train', transform=None, target_transform=None, download=False)

Bases: GTSRB

Custom GTSRB dataset that returns both augmented and non-augmented images.

class datasets.seq_gtrsrb.SequentialGTSRB(args, transform_type='weak')

Bases: ContinualDataset

The Sequential GTSRB dataset with 224x224 resolution with ViT-B/16.

MEAN = (0.48145466, 0.4578275, 0.40821073)

NAME: str = 'seq-gtsrb'

N_CLASSES: int = 43

N_CLASSES_PER_TASK: int = [10, 10, 10, 10, 3]

N_TASKS: int = 5

SETTING: str = 'class-il'

SIZE: Tuple[int] = (224, 224)

STD = (0.26862954, 0.26130258, 0.27577711)

TEST_TRANSFORM = Compose(     Resize(size=256, interpolation=bilinear, max_size=None, antialias=True)     CenterCrop(size=(224, 224))     ToTensor()     Normalize(mean=(0.48145466, 0.4578275, 0.40821073), std=(0.26862954, 0.26130258, 0.27577711)) )

TRANSFORM = Compose(     RandomResizedCrop(size=(224, 224), scale=(0.8, 1.0), ratio=(0.75, 1.3333), interpolation=bilinear, antialias=True)     RandomHorizontalFlip(p=0.5)     ColorJitter(brightness=(0.8, 1.2), contrast=(0.8, 1.2), saturation=(0.8, 1.2), hue=(-0.1, 0.1))     ToTensor()     Normalize(mean=(0.48145466, 0.4578275, 0.40821073), std=(0.26862954, 0.26130258, 0.27577711)) )

get_backbone()

get_batch_size()

get_class_names()

get_data_loaders()

Return type:

Tuple[DataLoader, DataLoader]

static get_denormalization_transform()

get_epochs()

static get_loss()

static get_normalization_transform()

static get_prompt_templates()

static get_transform()