DevMesh

Evo Norm layers in TensorFlow 2

Sayak Paul

Sayak Paul

Kolkata, West Bengal

0 0
  • 0 Collaborators

Presents implementations of EvoNormB0 and EvoNormS0 layers as proposed in Evolving Normalization-Activation Layers by Liu et al. The authors showed the results with these layers tested on MobileNetV2, ResNets, MnasNet, and EfficientNets. However, I tried a Mini Inception architecture. ...learn more

Project status: Under Development

Artificial Intelligence

Intel Technologies
Intel Python

Code Samples [1]

Overview / Usage

Presents implementations of EvoNormB0 and EvoNormS0 layers as proposed in Evolving Normalization-Activation Layers by Liu et al. The authors showed the results with these layers tested on MobileNetV2, ResNets, MnasNet, and EfficientNets. However, I tried a Mini Inception architecture as shown in this blog post with the CIFAR10 dataset.

Methodology / Approach

Instead of trying it out on MobileNetV2, ResNet, or EfficientNet I tried it on a *Mini Inception* architecture with CIFAR10 as the dataset. Additionally, I ran a mini hyperparameter search on the groups hyperparameter of the EvoNormS0 layer.

Technologies Used

  • TensorFlow

Repository

https://github.com/sayakpaul/EvoNorms-in-TensorFlow-2

Comments (0)