Tweeted By @ak92501
Searching for Efficient Multi-Stage Vision Transformers
— AK (@ak92501) September 3, 2021
abs: https://t.co/anNr3DoY74
Experiments on ImageNet demonstrate that ViT-ResNAS achieves better accuracy-MACs and accuracy-throughput trade-offs than the original DeiT and other strong baselines of ViT pic.twitter.com/iog50Sqi4B