RepMLP: Re-parameterizing Convolutions into Fully-connected Layers for Image Recognition

标题 RepMLP: Re-parameterizing Convolutions into Fully-connected Layers for Image Recognition 年份： 2021 年 5 月 GB/T 7714: Ding X, Zhang X, Han J, et al. RepMLP: Re-parameterizing Convolutions into Fully-connected Layers for Image Recognition[J]. arXiv preprint arXiv:2105.01883, 2021. 本文是清华大学&旷视科技在结构重参数领域继ACNe...