ResNet

Deep Residual Learning for Image Recognition

Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun · CVPR 2016

Read paper · arXiv:1512.03385

Implement a two-layer residual branch, identity or projection shortcut, and post-addition ReLU.

What you will implement

Complete one matrix residual block; a four-block synthetic regression experiment is supplied. Convolutions, BN and full ResNet training are outside scope.

• 97. ResNet: Residual Branches and Projection Shortcuts

  Implement a two-layer residual branch, identity or projection shortcut, and post-addition ReLU.

  Complete one matrix residual block; a four-block synthetic regression experiment is supplied. Convolutions, BN and full ResNet training are outside scope.