undo_affine_transform_batch

lightning_pose.data.utils.undo_affine_transform_batch(keypoints_augmented: ~torch.Annotated[~torch.Tensor, {'__torchtyping__': True, 'details': ('seq_len', 'num_keypointsx2',), 'cls_name': 'TensorType'}], transforms: ~torch.Annotated[~torch.Tensor, {'__torchtyping__': True, 'details': ('seq_len', h: 2, w: 3,), 'cls_name': 'TensorType'}] | ~torch.Annotated[~torch.Tensor, {'__torchtyping__': True, 'details': (h: 2, w: 3,), 'cls_name': 'TensorType'}] | ~torch.Annotated[~torch.Tensor, {'__torchtyping__': True, 'details': ('seq_len', null: 1,), 'cls_name': 'TensorType'}] | ~torch.Annotated[~torch.Tensor, {'__torchtyping__': True, 'details': (null: 1,), 'cls_name': 'TensorType'}] | ~torch.Annotated[~torch.Tensor, {'__torchtyping__': True, 'details': ('num_views', h: 2, w: 3,), 'cls_name': 'TensorType'}] | ~torch.Annotated[~torch.Tensor, {'__torchtyping__': True, 'details': ('num_views', null: 1, null: 1,), 'cls_name': 'TensorType'}], is_multiview: bool = False) → Tensor, {'__torchtyping__': True, 'details': ('seq_len', 'num_keypointsx2',), 'cls_name': 'TensorType'}]

Potentially undo an affine transform given a tensor of keypoints and the tranform matrix.