[Glean] ResNet-50 Architecture and # MACs

Published: by Creative Commons Licence (Last updated: )

ResNet-50 Architecture and # MACs

ResNet-50 Architecture1

Architectures for ResNet

From the figure above, ResNet-50 contains 2 separate convolutional layers plus 16 building block where each building block contains three convolutional layers.

Building Block1

Residual Learning building block

The building block in residual learning contains one residual representations and one shortcut connections which skipping one or more layers.

In ResNet-50, the shortcut connections skipping three layers.

# Weights and # MACs2

Number of Weights and MACs

We can conclude that the last three building blocks occupy more weights while their computation operations near the same between different building blocks.

  1. K. He, X. Zhang, S. Ren, and J. Sun, “Deep Residual Learning for Image Recognition,” arXiv:1512.03385 [cs], Dec. 2015, Accessed: Dec. 22, 2019. [Online]. Available: http://arxiv.org/abs/1512.03385.  2

  2. 经典神经网络参数的计算【不定期更新】, https://zhuanlan.zhihu.com/p/49842046