sketchgraphs_models.training.TrainingConfig

class sketchgraphs_models.training.TrainingConfig(dataloader: torch.utils.data.dataloader.DataLoader, tb_writer: Optional[torch.utils.tensorboard.writer.SummaryWriter], device: torch.device, batch_size: int, batches_per_epoch: Optional[int] = None)

Named tuple holding configuration for training a given model.

__getnewargs__()

Return self as a plain tuple. Used by copy and pickle.

static __new__(_cls, dataloader: torch.utils.data.dataloader.DataLoader, tb_writer: Optional[torch.utils.tensorboard.writer.SummaryWriter], device: torch.device, batch_size: int, batches_per_epoch: Optional[int] = None)

Create new instance of TrainingConfig(dataloader, tb_writer, device, batch_size, batches_per_epoch)

batch_size: int

Alias for field number 3

batches_per_epoch: Optional[int]

Alias for field number 4

dataloader: torch.utils.data.dataloader.DataLoader

Alias for field number 0

device: torch.device

Alias for field number 2

tb_writer: Optional[torch.utils.tensorboard.writer.SummaryWriter]

Alias for field number 1