ICARL CSCCT#

Arguments#

Options

--csc_weightfloat

Help: Weight of CSC loss.

  • Default: 3

--ct_weightfloat

Help: Weight of CT loss.

  • Default: 1.5

--ct_temperaturefloat

Help: Temperature of CT loss.

  • Default: 2

Rehearsal arguments

Arguments shared by all rehearsal-based methods.

--buffer_sizeint

Help: The size of the memory buffer.

  • Default: None

--minibatch_sizeint

Help: The batch size of the memory buffer.

  • Default: None

Classes#

class models.icarl_cscct.ICarlCscCt(backbone, loss, args, transform, dataset=None)[source]#

Bases: CscCtModel

Continual Learning via iCaRL. Treated with CSCCT!

COMPATIBILITY: List[str] = ['class-il', 'task-il']#
NAME: str = 'icarl_cscct'#
begin_task(dataset)[source]#
static binary_cross_entropy(pred, y)[source]#
compute_class_means()[source]#

Computes a vector representing mean features for each class.

end_task(dataset)[source]#
forward(x)[source]#
get_loss(inputs, labels, task_idx, logits)[source]#

Computes the loss tensor.

Parameters:

  • inputs (Tensor) – the images to be fed to the network

  • labels (Tensor) – the ground-truth labels

  • task_idx (int) – the task index

  • logits (Tensor) – the logits of the old network

Returns:

the differentiable loss value

Return type:

Tensor

static get_parser(parser)[source]#
Return type:

ArgumentParser

observe(inputs, labels, not_aug_inputs, logits=None, epoch=None)[source]#