class

CrossViT

extendsPretrainedModelBackboneMixin

CrossViT(config: CrossViTConfig)

source edit

CrossViT backbone (Chen et al., ICCV 2021).

Builds two parallel ViT branches that operate on different resolutions and exchange information through K=3 cross-attention fusion blocks. See CrossViTConfig for the per-variant hyperparameters.

Used by 2

Constructors

dunder

init

→None

__init__(config: CrossViTConfig)

source edit

Properties

prop

feature_info

→list[FeatureInfo]

feature_info: list[FeatureInfo]

source edit

Instance methods

forward

→BaseModelOutput

forward(x: Tensor)

source edit

forward_features

→tuple[Tensor, Tensor]

forward_features(x: Tensor)

source edit

class

CrossViT

extendsPretrainedModelBackboneMixin

CrossViT(config: CrossViTConfig)

source edit

CrossViT backbone (Chen et al., ICCV 2021).

Builds two parallel ViT branches that operate on different resolutions and exchange information through K=3 cross-attention fusion blocks. See CrossViTConfig for the per-variant hyperparameters.