avalanche.evaluation.metrics.AccuracyMatrixPluginMetric

class avalanche.evaluation.metrics.AccuracyMatrixPluginMetric[source]

Class for obtaining an Accuracy Matrix for the evaluation stream

__init__()[source]

Creates the Accuracy Matrix plugin

Methods

__init__()

Creates the Accuracy Matrix plugin

add_new_task(new_length)

Adds a new dimension to the accuracy matrix.

after_backward(strategy)

after_eval(strategy)

Performs actions after the evaluation phase.

after_eval_dataset_adaptation(strategy)

after_eval_exp(strategy)

Performs actions after evaluating an experience.

after_eval_forward(strategy)

after_eval_iteration(strategy)

Performs actions after each evaluation iteration.

after_forward(strategy)

after_train_dataset_adaptation(strategy)

after_training(strategy)

after_training_epoch(strategy)

after_training_exp(strategy)

after_training_iteration(strategy)

after_update(strategy)

before_backward(strategy)

before_eval(strategy)

before_eval_dataset_adaptation(strategy)

before_eval_exp(strategy)

before_eval_forward(strategy)

before_eval_iteration(strategy)

before_forward(strategy)

before_train_dataset_adaptation(strategy)

before_training(strategy)

Performs actions before the training phase.

before_training_epoch(strategy)

before_training_exp(strategy)

before_training_iteration(strategy)

before_update(strategy)

reset([strategy])

Resets the metric.

result([strategy])

Returns the metric result.

update(num_training_steps, eval_exp_id)

Updates the matrix with the accuracy value for a given task pair.