Skip to content

mniyas/mlp-mixer

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 

Repository files navigation

mlp-mixer

This is an adaptation of MLP-Mixer: An all-MLP Architecture for Vision on MNIST-Fashion dataset.

MLP-Mixer Architecture

Convolutional Neural Networks (CNNs) are the go-to model for computer vision. Recently, attention-based networks, such as the Vision Transformer, have also become popular. MLP-Mixer architecture show that while convolutions and attention are both sufficient for good performance, neither of them are necessary. MLP-Mixer contains two types of layers: one with MLPs applied independently to image patches (i.e. "mixing" the per-location features), and one with MLPs applied across patches (i.e. "mixing" spatial information).

This model achieves a classification accuracy of 90.2% on MNIST-Fashion dataset with 75 epochs. Model could be improved further by adding depth and residual connections.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published