SwAV

class mmpretrain.models.selfsup.SwAV(backbone, neck=None, head=None, target_generator=None, pretrained=None, data_preprocessor=None, init_cfg=None)

SwAV.

Implementation of Unsupervised Learning of Visual Features by Contrasting Cluster Assignments.

The queue is built in mmpretrain/engine/hooks/swav_hook.py.

loss(inputs, data_samples, **kwargs)

Forward computation during training.

Parameters:

• inputs (List[torch.Tensor]) – The input images.

• data_samples (List[DataSample]) – All elements required during the forward function.

Returns:

A dictionary of loss components.

Return type:

Dict[str, torch.Tensor]