nntoolbox.transforms.mixup module

class nntoolbox.transforms.mixup.MixupTransformer(alpha: float = 0.4)

Bases: object

Implement mixup data augmentation:

x = tau x_1 + (1 - tau) x_2

y = tau y_1 + (1 - tau) y_2

References:

https://arxiv.org/pdf/1710.09412.pdf

Based on fastai implementation: https://github.com/fastai/fastai/blob/master/fastai/callbacks/mixup.py

transform_data(images: torch.Tensor, labels: torch.Tensor) → Tuple[torch.Tensor, torch.Tensor]

transform_loss(criterion, train)