avalanche.training.ExpertGateStrategy

class avalanche.training.ExpertGateStrategy(*, model: ~torch.nn.modules.module.Module, optimizer: ~torch.optim.optimizer.Optimizer, criterion: ~torch.nn.modules.module.Module | ~typing.Callable[[~torch.Tensor, ~torch.Tensor], ~torch.Tensor] = CrossEntropyLoss(), train_mb_size: int = 1, train_epochs: int = 1, eval_mb_size: int | None = None, device=None, plugins: ~typing.List[~avalanche.core.SupervisedPlugin] | None = None, evaluator: ~avalanche.training.plugins.evaluation.EvaluationPlugin = <function default_evaluator>, eval_every=-1, ae_train_mb_size=1, ae_train_epochs=2, ae_latent_dim=100, ae_lr=0.001, temp=2, rel_thresh=0.85, **base_kwargs)[source]

Expert Gate strategy. New experts are trained and added to the model as tasks are learned sequentially.

Technique introduced in: ‘Aljundi, Rahaf, Punarjay Chakravarty, and Tinne Tuytelaars. “Expert gate: Lifelong learning with a network of experts.” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017.’ https://arxiv.org/abs/1611.06194

To use this strategy you need to instantiate an ExpertGate model. See the ExpertGate plugin for more details.

__init__(*, model: ~torch.nn.modules.module.Module, optimizer: ~torch.optim.optimizer.Optimizer, criterion: ~torch.nn.modules.module.Module | ~typing.Callable[[~torch.Tensor, ~torch.Tensor], ~torch.Tensor] = CrossEntropyLoss(), train_mb_size: int = 1, train_epochs: int = 1, eval_mb_size: int | None = None, device=None, plugins: ~typing.List[~avalanche.core.SupervisedPlugin] | None = None, evaluator: ~avalanche.training.plugins.evaluation.EvaluationPlugin = <function default_evaluator>, eval_every=-1, ae_train_mb_size=1, ae_train_epochs=2, ae_latent_dim=100, ae_lr=0.001, temp=2, rel_thresh=0.85, **base_kwargs)[source]

Init.

Parameters:
  • model – PyTorch model.

  • optimizer – PyTorch optimizer.

  • criterion – loss function.

  • train_mb_size – mini-batch size for training.

  • train_epochs – number of training epochs.

  • eval_mb_size – mini-batch size for eval.

  • device – PyTorch device where the model will be allocated.

  • plugins – (optional) list of StrategyPlugins.

  • evaluator – (optional) instance of EvaluationPlugin for logging and metric computations. None to remove logging.

  • eval_every – the frequency of the calls to eval inside the training loop. -1 disables the evaluation. 0 means eval is called only at the end of the learning experience. Values >0 mean that eval is called every eval_every epochs and at the end of the learning experience.

  • base_kwargs – any additional

  • ae_train_mb_size – mini-batch size for training of the autoencoder

  • ae_train_epochs – number of training epochs for the autoencoder

  • ae_lr – the learning rate for the autoencoder training

using vanilla SGD :param temp: the temperature hyperparameter when selecting the expert during the forward method

BaseTemplate constructor arguments.

Methods

__init__(*, model, optimizer[, criterion, ...])

Init.

backward()

Run the backward pass.

check_model_and_optimizer([...])

criterion()

Loss function for supervised problems.

eval(exp_list, **kwargs)

Evaluate the current model on a series of experiences and returns the last recorded value for each metric.

eval_dataset_adaptation(**kwargs)

Initialize self.adapted_dataset.

eval_epoch(**kwargs)

Evaluation loop over the current self.dataloader.

forward()

Compute the model's output given the current mini-batch.

make_eval_dataloader([num_workers, shuffle, ...])

Initializes the eval data loader. :param num_workers: How many subprocesses to use for data loading. 0 means that the data will be loaded in the main process. (default: 0). :param pin_memory: If True, the data loader will copy Tensors into CUDA pinned memory before returning them. Defaults to True. :param kwargs: :return:.

make_optimizer([reset_optimizer_state, ...])

Optimizer initialization.

make_train_dataloader([num_workers, ...])

Data loader initialization.

model_adaptation([model])

Adapts the model to the current data.

optimizer_step()

Execute the optimizer step (weights update).

stop_training()

Signals to stop training at the next iteration.

train(experiences[, eval_streams])

Training loop.

train_dataset_adaptation(**kwargs)

Initialize self.adapted_dataset.

training_epoch(**kwargs)

Training epoch.

Attributes

is_eval

True if the strategy is in evaluation mode.

mb_task_id

Current mini-batch task labels.

mb_x

Current mini-batch input.

mb_y

Current mini-batch target.

mbatch

Current mini-batch.

mb_output

Model's output computed on the current mini-batch.

dataloader

Dataloader.

optimizer

PyTorch optimizer.

loss

Loss of the current mini-batch.

train_epochs

Number of training epochs.

train_mb_size

Training mini-batch size.

eval_mb_size

Eval mini-batch size.

retain_graph

Retain graph when calling loss.backward().

evaluator

EvaluationPlugin used for logging and metric computations.

clock

Incremental counters for strategy events.

adapted_dataset

Data used to train.

model

PyTorch model.

device

PyTorch device where the model will be allocated.

plugins

List of `SupervisedPlugin`s. .

experience

Current experience.

is_training

True if the strategy is in training mode.

current_eval_stream

Current evaluation stream.