MILAN¶
- class mmpretrain.models.selfsup.MILAN(backbone, neck=None, head=None, target_generator=None, pretrained=None, data_preprocessor=None, init_cfg=None)[源代码]¶
MILAN.
Implementation of MILAN: Masked Image Pretraining on Language Assisted Representation.
- loss(inputs, data_samples, **kwargs)[源代码]¶
The forward function in training.
- 参数:
inputs (torch.Tensor) – The input images.
data_samples (List[DataSample]) – All elements required during the forward function.
- 返回:
A dictionary of loss components.
- 返回类型:
Dict[str, torch.Tensor]