Contextual Transformer Networks for Visual Recognition论文以及代码解析

Contextual Transformer Networks for Visual Recognition

  • 1. Abstract
  • 2. Introduction
  • 3. Approach
    • 3.1. Multi-head Self-attention in Vision Backbones
    • 3.2. Contextual Transformer Block
    • 3.3. Contextual Transformer Networks
  • 4. 代码解析
    • 4.1. train脚本中参数的配置
    • 4.2. model脚本
      • 4.2.1. 整体forward过程

论文地址: CoTNet
源码地址:

你可能感兴趣的:(transformer,深度学习,计算机视觉,python,pytorch)