VisionTransformer¶

模型構建器¶

以下模型構建器可用於例項化 VisionTransformer 模型，可選擇是否載入預訓練權重。所有模型構建器均在內部依賴 torchvision.models.vision_transformer.VisionTransformer 基類。有關此類更多詳細資訊，請參閱原始碼。

`vit_b_16`(*[, weights, progress])	從 An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale 構建 vit_b_16 架構。
`vit_b_32`(*[, weights, progress])	從 An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale 構建 vit_b_32 架構。
`vit_l_16`(*[, weights, progress])	從 An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale 構建 vit_l_16 架構。
`vit_l_32`(*[, weights, progress])	從 An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale 構建 vit_l_32 架構。
`vit_h_14`(*[, weights, progress])	從 An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale 構建 vit_h_14 架構。