avalanche.training.LearningToPrompt

class avalanche.training.LearningToPrompt(*, model_name: str, criterion: ~torch.nn.modules.module.Module = CrossEntropyLoss(), train_mb_size: int = 1, train_epochs: int = 1, eval_mb_size: int | None = 1, device: str | ~torch.device = 'cpu', plugins: ~typing.List[~avalanche.core.SupervisedPlugin] | None = None, evaluator: ~avalanche.training.plugins.evaluation.EvaluationPlugin | ~typing.Callable[[], ~avalanche.training.plugins.evaluation.EvaluationPlugin] = <function default_evaluator>, eval_every: int = -1, peval_mode: str = 'epoch', prompt_pool: bool = True, pool_size: int = 20, prompt_length: int = 5, top_k: int = 5, lr: float = 0.03, sim_coefficient: float = 0.1, prompt_key: bool = True, pretrained: bool = True, num_classes: int = 10, drop_rate: float = 0.0, drop_path_rate: float = 0.0, embedding_key: str = 'cls', prompt_init: str = 'uniform', batchwise_prompt: bool = False, head_type: str = 'prompt', use_prompt_mask: bool = False, train_prompt_mask: bool = False, use_cls_features: bool = True, use_mask: bool = True, use_vit: bool = True, **kwargs)

Learning to Prompt (L2P) strategy.

Technique introduced in: “Wang, Zifeng, et al. “Learning to prompt for continual learning.” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2022.”

Implementation based on: - https://github.com/JH-LEE-KR/l2p-pytorch - And implementations by Dario Salvati

As a model_name, we expect to receive one of the model list in avalanche.models.vit

Those models are based on the library timm.

__init__(*, model_name: str, criterion: ~torch.nn.modules.module.Module = CrossEntropyLoss(), train_mb_size: int = 1, train_epochs: int = 1, eval_mb_size: int | None = 1, device: str | ~torch.device = 'cpu', plugins: ~typing.List[~avalanche.core.SupervisedPlugin] | None = None, evaluator: ~avalanche.training.plugins.evaluation.EvaluationPlugin | ~typing.Callable[[], ~avalanche.training.plugins.evaluation.EvaluationPlugin] = <function default_evaluator>, eval_every: int = -1, peval_mode: str = 'epoch', prompt_pool: bool = True, pool_size: int = 20, prompt_length: int = 5, top_k: int = 5, lr: float = 0.03, sim_coefficient: float = 0.1, prompt_key: bool = True, pretrained: bool = True, num_classes: int = 10, drop_rate: float = 0.0, drop_path_rate: float = 0.0, embedding_key: str = 'cls', prompt_init: str = 'uniform', batchwise_prompt: bool = False, head_type: str = 'prompt', use_prompt_mask: bool = False, train_prompt_mask: bool = False, use_cls_features: bool = True, use_mask: bool = True, use_vit: bool = True, **kwargs)

Init.

Parameters:

• model_name – Name of the model to use. For a complete list check models.vit.py

• criterion – Loss functions used during training. Default CrossEntropyLoss.

• train_mb_size – The train minibatch size. Defaults to 1.

• train_epochs – The number of training epochs. Defaults to 1.

• eval_mb_size – The eval minibatch size. Defaults to 1.

• device – The device to use. Defaults to None (cpu).

• plugins – Plugins to be added. Defaults to None.

• evaluator – (optional) instance of EvaluationPlugin for logging and metric computations.

• 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.

• use_cls_features – Use an external pre-trained model to obtained features to obtained the prompts.

• use_mask – Use mask to train only classification rows of the classes of the current task. Default True.

• use_vit – Boolean to confirm the usage of a visual Transformer. Default True

Methods

__init__(*, model_name[, 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.