TWP

TWP is a continual learning method for graphs, which stabilizes parameters important in topological aggregation by graph neural networks through regularization. For the details, see the original paper.

Node-level Problems

class NCClassILTWPMinibatchTrainer(model, scenario, optimizer_fn, loss_fn, device, **kwargs)

afterInference(results, model, optimizer, _curr_batch, training_states)

The event function to execute some processes right after the inference step (for training). We recommend performing backpropagation in this event function.

TWP performs regularization process in this function.

Parameters

• results (dict) – the returned dictionary from the event function inference.

• model (torch.nn.Module) – the current trained model.

• optimizer (torch.optim.Optimizer) – the current optimizer function.

• curr_batch (object) – the data (or minibatch) for the current iteration.

• curr_training_states (dict) – the dictionary containing the current training states.

Returns

A dictionary containing the information from the results.

inference(model, _curr_batch, training_states, return_elist=False)

The event function to execute inference step.

TWP requires edge weights computed by attention mechanism.

Parameters

• model (torch.nn.Module) – the current trained model.

• curr_batch (object) – the data (or minibatch) for the current iteration.

• curr_training_states (dict) – the dictionary containing the current training states.

Returns

A dictionary containing the inference results, such as prediction result and loss.

initTrainingStates(scenario, model, optimizer)

The event function to initialize the dictionary for storing training states (i.e., intermedeiate results).

Parameters

• scenario (begin.scenarios.common.BaseScenarioLoader) – the given ScenarioLoader to the trainer

• model (torch.nn.Module) – the given model to the trainer

• optmizer (torch.optim.Optimizer) – the optimizer generated from the given optimizer_fn

Returns

Initialized training state (dict).

processAfterTraining(task_id, curr_dataset, curr_model, curr_optimizer, curr_training_states)

The event function to execute some processes after training the current task.

TWP computes weights for regularization process and stores the learned weights in this function.

Parameters

• task_id (int) – the index of the current task.

• curr_dataset (object) – The dataset for the current task.

• curr_model (torch.nn.Module) – the current trained model.

• curr_optimizer (torch.optim.Optimizer) – the current optimizer function.

• curr_training_states (dict) – the dictionary containing the current training states.

class NCClassILTWPTrainer(model, scenario, optimizer_fn, loss_fn, device, **kwargs)

afterInference(results, model, optimizer, _curr_batch, training_states)

The event function to execute some processes right after the inference step (for training). We recommend performing backpropagation in this event function.

TWP performs regularization process in this function.

Parameters

• results (dict) – the returned dictionary from the event function inference.

• model (torch.nn.Module) – the current trained model.

• optimizer (torch.optim.Optimizer) – the current optimizer function.

• curr_batch (object) – the data (or minibatch) for the current iteration.

• curr_training_states (dict) – the dictionary containing the current training states.

Returns

A dictionary containing the information from the results.

inference(model, _curr_batch, training_states, return_elist=False)

The event function to execute inference step.

TWP requires edge weights computed by attention mechanism.

Parameters

• model (torch.nn.Module) – the current trained model.

• curr_batch (object) – the data (or minibatch) for the current iteration.

• curr_training_states (dict) – the dictionary containing the current training states.

Returns

A dictionary containing the inference results, such as prediction result and loss.

initTrainingStates(scenario, model, optimizer)

The event function to initialize the dictionary for storing training states (i.e., intermedeiate results).

Parameters

• scenario (begin.scenarios.common.BaseScenarioLoader) – the given ScenarioLoader to the trainer

• model (torch.nn.Module) – the given model to the trainer

• optmizer (torch.optim.Optimizer) – the optimizer generated from the given optimizer_fn

Returns

Initialized training state (dict).

processAfterTraining(task_id, curr_dataset, curr_model, curr_optimizer, curr_training_states)

The event function to execute some processes after training the current task.

TWP computes weights for regularization process and stores the learned weights in this function.

Parameters

• task_id (int) – the index of the current task.

• curr_dataset (object) – The dataset for the current task.

• curr_model (torch.nn.Module) – the current trained model.

• curr_optimizer (torch.optim.Optimizer) – the current optimizer function.

• curr_training_states (dict) – the dictionary containing the current training states.

class NCDomainILTWPTrainer(model, scenario, optimizer_fn, loss_fn, device, **kwargs)

This trainer has the same behavior as NCClassILTWPTrainer.

class NCTaskILTWPTrainer(model, scenario, optimizer_fn, loss_fn, device, **kwargs)

afterInference(results, model, optimizer, _curr_batch, training_states)

The event function to execute some processes right after the inference step (for training). We recommend performing backpropagation in this event function.

TWP performs regularization process in this function.

Parameters

• results (dict) – the returned dictionary from the event function inference.

• model (torch.nn.Module) – the current trained model.

• optimizer (torch.optim.Optimizer) – the current optimizer function.

• curr_batch (object) – the data (or minibatch) for the current iteration.

• curr_training_states (dict) – the dictionary containing the current training states.

Returns

A dictionary containing the information from the results.

inference(model, _curr_batch, training_states, return_elist=False)

The event function to execute inference step.

For task-IL, we need to additionally consider task information for the inference step. TWP requires edge weights computed by attention mechanism.

Parameters

• model (torch.nn.Module) – the current trained model.

• curr_batch (object) – the data (or minibatch) for the current iteration.

• curr_training_states (dict) – the dictionary containing the current training states.

Returns

A dictionary containing the inference results, such as prediction result and loss.

initTrainingStates(scenario, model, optimizer)

The event function to initialize the dictionary for storing training states (i.e., intermedeiate results).

Parameters

• scenario (begin.scenarios.common.BaseScenarioLoader) – the given ScenarioLoader to the trainer

• model (torch.nn.Module) – the given model to the trainer

• optmizer (torch.optim.Optimizer) – the optimizer generated from the given optimizer_fn

Returns

Initialized training state (dict).

processAfterTraining(task_id, curr_dataset, curr_model, curr_optimizer, curr_training_states)

The event function to execute some processes after training the current task.

TWP computes weights for regularization process and stores the learned weights in this function.

Parameters

• task_id (int) – the index of the current task.

• curr_dataset (object) – The dataset for the current task.

• curr_model (torch.nn.Module) – the current trained model.

• curr_optimizer (torch.optim.Optimizer) – the current optimizer function.

• curr_training_states (dict) – the dictionary containing the current training states.

class NCTimeILTWPTrainer(model, scenario, optimizer_fn, loss_fn, device, **kwargs)

This trainer has the same behavior as NCClassILTWPTrainer.

Link-level Problems

class LCClassILTWPTrainer(model, scenario, optimizer_fn, loss_fn, device, **kwargs)

afterInference(results, model, optimizer, _curr_batch, training_states)

The event function to execute some processes right after the inference step (for training). We recommend performing backpropagation in this event function.

TWP performs regularization process in this function.

Parameters

• results (dict) – the returned dictionary from the event function inference.

• model (torch.nn.Module) – the current trained model.

• optimizer (torch.optim.Optimizer) – the current optimizer function.

• curr_batch (object) – the data (or minibatch) for the current iteration.

• curr_training_states (dict) – the dictionary containing the current training states.

Returns

A dictionary containing the information from the results.

inference(model, _curr_batch, training_states, return_elist=False)

The event function to execute inference step.

TWP requires edge weights computed by attention mechanism.

Parameters

• model (torch.nn.Module) – the current trained model.

• curr_batch (object) – the data (or minibatch) for the current iteration.

• curr_training_states (dict) – the dictionary containing the current training states.

Returns

A dictionary containing the inference results, such as prediction result and loss.

initTrainingStates(scenario, model, optimizer)

The event function to initialize the dictionary for storing training states (i.e., intermedeiate results).

Parameters

• scenario (begin.scenarios.common.BaseScenarioLoader) – the given ScenarioLoader to the trainer

• model (torch.nn.Module) – the given model to the trainer

• optmizer (torch.optim.Optimizer) – the optimizer generated from the given optimizer_fn

Returns

Initialized training state (dict).

processAfterTraining(task_id, curr_dataset, curr_model, curr_optimizer, curr_training_states)

The event function to execute some processes after training the current task.

TWP computes weights for regularization process and stores the learned weights in this function.

Parameters

• task_id (int) – the index of the current task.

• curr_dataset (object) – The dataset for the current task.

• curr_model (torch.nn.Module) – the current trained model.

• curr_optimizer (torch.optim.Optimizer) – the current optimizer function.

• curr_training_states (dict) – the dictionary containing the current training states.

class LCTaskILTWPTrainer(model, scenario, optimizer_fn, loss_fn, device, **kwargs)

afterInference(results, model, optimizer, _curr_batch, training_states)

The event function to execute some processes right after the inference step (for training). We recommend performing backpropagation in this event function.

TWP performs regularization process in this function.

Parameters

• results (dict) – the returned dictionary from the event function inference.

• model (torch.nn.Module) – the current trained model.

• optimizer (torch.optim.Optimizer) – the current optimizer function.

• curr_batch (object) – the data (or minibatch) for the current iteration.

• curr_training_states (dict) – the dictionary containing the current training states.

Returns

A dictionary containing the information from the results.

inference(model, _curr_batch, training_states, return_elist=False)

The event function to execute inference step.

For task-IL, we need to additionally consider task information for the inference step. TWP requires edge weights computed by attention mechanism.

Parameters

• model (torch.nn.Module) – the current trained model.

• curr_batch (object) – the data (or minibatch) for the current iteration.

• curr_training_states (dict) – the dictionary containing the current training states.

Returns

A dictionary containing the inference results, such as prediction result and loss.

initTrainingStates(scenario, model, optimizer)

The event function to initialize the dictionary for storing training states (i.e., intermedeiate results).

Parameters

• scenario (begin.scenarios.common.BaseScenarioLoader) – the given ScenarioLoader to the trainer

• model (torch.nn.Module) – the given model to the trainer

• optmizer (torch.optim.Optimizer) – the optimizer generated from the given optimizer_fn

Returns

Initialized training state (dict).

prepareLoader(_curr_dataset, curr_training_states)

The event function to generate dataloaders from the given dataset for the current task.

For task-IL, we need to additionally consider task information for the inference step.

Parameters

• curr_dataset (object) – The dataset for the current task. Its type is dgl.graph for node-level and link-level problem, and dgl.data.DGLDataset for graph-level problem.

• curr_training_states (dict) – the dictionary containing the current training states.

Returns

A tuple containing three dataloaders. The trainer considers the first dataloader, second dataloader, and third dataloader as dataloaders for training, validation, and test, respectively.

processAfterTraining(task_id, curr_dataset, curr_model, curr_optimizer, curr_training_states)

The event function to execute some processes after training the current task.

TWP computes weights for regularization process and stores the learned weights in this function.

Parameters

• task_id (int) – the index of the current task.

• curr_dataset (object) – The dataset for the current task.

• curr_model (torch.nn.Module) – the current trained model.

• curr_optimizer (torch.optim.Optimizer) – the current optimizer function.

• curr_training_states (dict) – the dictionary containing the current training states.

class LCTimeILTWPTrainer(model, scenario, optimizer_fn, loss_fn, device, **kwargs)

afterInference(results, model, optimizer, _curr_batch, training_states)

The event function to execute some processes right after the inference step (for training). We recommend performing backpropagation in this event function.

TWP performs regularization process in this function.

Parameters

• results (dict) – the returned dictionary from the event function inference.

• model (torch.nn.Module) – the current trained model.

• optimizer (torch.optim.Optimizer) – the current optimizer function.

• curr_batch (object) – the data (or minibatch) for the current iteration.

• curr_training_states (dict) – the dictionary containing the current training states.

Returns

A dictionary containing the information from the results.

inference(model, _curr_batch, training_states, return_elist=False)

The event function to execute inference step.

TWP requires edge weights computed by attention mechanism.

Parameters

• model (torch.nn.Module) – the current trained model.

• curr_batch (object) – the data (or minibatch) for the current iteration.

• curr_training_states (dict) – the dictionary containing the current training states.

Returns

A dictionary containing the inference results, such as prediction result and loss.

initTrainingStates(scenario, model, optimizer)

The event function to initialize the dictionary for storing training states (i.e., intermedeiate results).

Parameters

• scenario (begin.scenarios.common.BaseScenarioLoader) – the given ScenarioLoader to the trainer

• model (torch.nn.Module) – the given model to the trainer

• optmizer (torch.optim.Optimizer) – the optimizer generated from the given optimizer_fn

Returns

Initialized training state (dict).

processAfterTraining(task_id, curr_dataset, curr_model, curr_optimizer, curr_training_states)

The event function to execute some processes after training the current task.

TWP computes weights for regularization process and stores the learned weights in this function.

Parameters

• task_id (int) – the index of the current task.

• curr_dataset (object) – The dataset for the current task.

• curr_model (torch.nn.Module) – the current trained model.

• curr_optimizer (torch.optim.Optimizer) – the current optimizer function.

• curr_training_states (dict) – the dictionary containing the current training states.

processBeforeTraining(task_id, curr_dataset, curr_model, curr_optimizer, curr_training_states)

The event function to execute some processes before training.

We need to extend the function since the output format is slightly different from the base trainer.

Parameters

• task_id (int) – the index of the current task

• curr_dataset (object) – The dataset for the current task.

• curr_model (torch.nn.Module) – the current trained model.

• curr_optimizer (torch.optim.Optimizer) – the current optimizer function.

• curr_training_states (dict) – the dictionary containing the current training states.

processEvalIteration(model, _curr_batch)

The event function to handle every evaluation iteration.

We need to extend the function since the output format is slightly different from the base trainer.

Parameters

• model (torch.nn.Module) – the current trained model.

• curr_batch (object) – the data (or minibatch) for the current iteration.

Returns

A dictionary containing the outcomes (stats) during the evaluation iteration.

class LPDomainILTWPTrainer(model, scenario, optimizer_fn, loss_fn, device, **kwargs)

This trainer has the same behavior as LPTimeILTWPTrainer.

class LPTimeILTWPTrainer(model, scenario, optimizer_fn, loss_fn, device, **kwargs)

initTrainingStates(scenario, model, optimizer)

The event function to initialize the dictionary for storing training states (i.e., intermedeiate results).

Parameters

• scenario (begin.scenarios.common.BaseScenarioLoader) – the given ScenarioLoader to the trainer

• model (torch.nn.Module) – the given model to the trainer

• optmizer (torch.optim.Optimizer) – the optimizer generated from the given optimizer_fn

Returns

Initialized training state (dict).

processAfterTraining(task_id, curr_dataset, curr_model, curr_optimizer, curr_training_states)

The event function to execute some processes after training the current task.

TWP computes weights for regularization process and stores the learned weights in this function.

Parameters

• task_id (int) – the index of the current task.

• curr_dataset (object) – The dataset for the current task.

• curr_model (torch.nn.Module) – the current trained model.

• curr_optimizer (torch.optim.Optimizer) – the current optimizer function.

• curr_training_states (dict) – the dictionary containing the current training states.

processTrainIteration(model, optimizer, _curr_batch, training_states)

The event function to handle every training iteration.

TWP performs inference and regularization process in this function.

Parameters

• model (torch.nn.Module) – the current trained model.

• optimizer (torch.optim.Optimizer) – the current optimizer function.

• curr_batch (object) – the data (or minibatch) for the current iteration.

• curr_training_states (dict) – the dictionary containing the current training states.

Returns

A dictionary containing the outcomes (stats) during the training iteration.

Graph-level Problems

class GCClassILTWPTrainer(model, scenario, optimizer_fn, loss_fn, device, **kwargs)

afterInference(results, model, optimizer, _curr_batch, training_states)

The event function to execute some processes right after the inference step (for training). We recommend performing backpropagation in this event function.

TWP performs regularization process in this function.

Parameters

• results (dict) – the returned dictionary from the event function inference.

• model (torch.nn.Module) – the current trained model.

• optimizer (torch.optim.Optimizer) – the current optimizer function.

• curr_batch (object) – the data (or minibatch) for the current iteration.

• curr_training_states (dict) – the dictionary containing the current training states.

Returns

A dictionary containing the information from the results.

inference(model, _curr_batch, training_states, return_elist=False)

The event function to execute inference step.

TWP requires edge weights computed by attention mechanism.

Parameters

• model (torch.nn.Module) – the current trained model.

• curr_batch (object) – the data (or minibatch) for the current iteration.

• curr_training_states (dict) – the dictionary containing the current training states.

Returns

A dictionary containing the inference results, such as prediction result and loss.

initTrainingStates(scenario, model, optimizer)

The event function to initialize the dictionary for storing training states (i.e., intermedeiate results).

Parameters

• scenario (begin.scenarios.common.BaseScenarioLoader) – the given ScenarioLoader to the trainer

• model (torch.nn.Module) – the given model to the trainer

• optmizer (torch.optim.Optimizer) – the optimizer generated from the given optimizer_fn

Returns

Initialized training state (dict).

processAfterTraining(task_id, curr_dataset, curr_model, curr_optimizer, curr_training_states)

The event function to execute some processes after training the current task.

TWP computes weights for regularization process and stores the learned weights in this function.

Parameters

• task_id (int) – the index of the current task.

• curr_dataset (object) – The dataset for the current task.

• curr_model (torch.nn.Module) – the current trained model.

• curr_optimizer (torch.optim.Optimizer) – the current optimizer function.

• curr_training_states (dict) – the dictionary containing the current training states.

class GCDomainILTWPTrainer(model, scenario, optimizer_fn, loss_fn, device, **kwargs)

This trainer has the same behavior as GCClassILTWPTrainer.

class GCTaskILTWPTrainer(model, scenario, optimizer_fn, loss_fn, device, **kwargs)

afterInference(results, model, optimizer, _curr_batch, training_states)

The event function to execute some processes right after the inference step (for training). We recommend performing backpropagation in this event function.

TWP performs regularization process in this function.

Parameters

• results (dict) – the returned dictionary from the event function inference.

• model (torch.nn.Module) – the current trained model.

• optimizer (torch.optim.Optimizer) – the current optimizer function.

• curr_batch (object) – the data (or minibatch) for the current iteration.

• curr_training_states (dict) – the dictionary containing the current training states.

Returns

A dictionary containing the information from the results.

inference(model, _curr_batch, training_states, return_elist=False)

The event function to execute inference step.

For task-IL, we need to additionally consider task information for the inference step. TWP requires edge weights computed by attention mechanism.

Parameters

• model (torch.nn.Module) – the current trained model.

• curr_batch (object) – the data (or minibatch) for the current iteration.

• curr_training_states (dict) – the dictionary containing the current training states.

Returns

A dictionary containing the inference results, such as prediction result and loss.

initTrainingStates(scenario, model, optimizer)

The event function to initialize the dictionary for storing training states (i.e., intermedeiate results).

Parameters

• scenario (begin.scenarios.common.BaseScenarioLoader) – the given ScenarioLoader to the trainer

• model (torch.nn.Module) – the given model to the trainer

• optmizer (torch.optim.Optimizer) – the optimizer generated from the given optimizer_fn

Returns

Initialized training state (dict).

processAfterTraining(task_id, curr_dataset, curr_model, curr_optimizer, curr_training_states)

The event function to execute some processes after training the current task.

TWP computes weights for regularization process and stores the learned weights in this function.

Parameters

• task_id (int) – the index of the current task.

• curr_dataset (object) – The dataset for the current task.

• curr_model (torch.nn.Module) – the current trained model.

• curr_optimizer (torch.optim.Optimizer) – the current optimizer function.

• curr_training_states (dict) – the dictionary containing the current training states.

class GCTimeILTWPTrainer(model, scenario, optimizer_fn, loss_fn, device, **kwargs)

This trainer has the same behavior as GCClassILTWPTrainer.