ER STAR

Arguments

Options

--p_stepsint

Help: None

• Default: 1

--p_lamfloat

Help: None

• Default: 0.01

--p_gammafloat

Help: how far we can go from original weights

• Default: 0.05

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

This module implements the simplest form of rehearsal training: Experience Replay. It maintains a buffer of previously seen examples and uses them to augment the current batch during training.

Example usage:

model = Er(backbone, loss, args, transform) loss = model.observe(inputs, labels, not_aug_inputs, epoch)

Classes

class models.er_star.ErSTAR(backbone, loss, args, transform, dataset=None)

Bases: ContinualModel

COMPATIBILITY: List[str] = ['class-il', 'domain-il', 'task-il', 'general-continual']

NAME: str = 'er_star'

static get_parser(parser)

Returns an ArgumentParser object with predefined arguments for the Er model.

Besides the required add_management_args and add_experiment_args, this model requires the add_rehearsal_args to include the buffer-related arguments.

Return type:

ArgumentParser

observe(inputs, labels, not_aug_inputs, epoch=None)

ER trains on the current task using the data provided, but also augments the batch with data from the buffer.