google-research-vision_transformer-143bd26:基于 JAX/Flax 的视觉Transformer与MLP-Mixer模型项目

ICLR ICLR 2021 Oral An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale