mixmo.learners.abstract_learner.AbstractLearner¶

class mixmo.learners.abstract_learner.AbstractLearner(config_args, dloader, device)[source]¶

Bases: object

Base learner class that groups models, optimizers and loggers Performs the entire model building, training and evaluating process

__init__(config_args, dloader, device)[source]¶: Initialize self. See help(type(self)) for accurate signature.

Methods

`__init__`(config_args, dloader, device)	Initialize self.
`evaluate`(inference_loader[, split])	Perform an evaluation of the model
`evaluate_loop`(dloader, verbose, **kwargs)
`load_checkpoint`(checkpoint[, …])	Load checkpoint (and optimizer if included) to the wrapped model
`save_checkpoint`(epoch[, save_path])	Save model (and optimizer) state dict
`save_tb`(logs_dict, epoch)	Write stats from logs_dict at epoch to the Tensoboard summary writer
`train`(epoch)	Train for one epoch
`train_loop`(epoch)

Attributes

Get (or initialize) the Tensorboard SummaryWriter

_create_model_wrapper()[source]¶: Initialize the model along with other elements through a ModelWrapper

evaluate(inference_loader, split='test')[source]¶: Perform an evaluation of the model

load_checkpoint(checkpoint, include_optimizer=True, return_epoch=False)[source]¶: Load checkpoint (and optimizer if included) to the wrapped model

save_checkpoint(epoch, save_path=None)[source]¶: Save model (and optimizer) state dict

save_tb(logs_dict, epoch)[source]¶: Write stats from logs_dict at epoch to the Tensoboard summary writer