An All-MLP solution for Vision, from Google AI

Last update: Jan 06, 2023

Related tags

Overview

MLP Mixer - Pytorch

An All-MLP solution for Vision, from Google AI, in Pytorch.

No convolutions nor attention needed!

Yannic Kilcher video

Install

$ pip install mlp-mixer-pytorch

Usage

import torch
from mlp_mixer_pytorch import MLPMixer

model = MLPMixer(
    image_size = 256,
    patch_size = 16,
    dim = 512,
    depth = 12,
    num_classes = 1000
)

img = torch.randn(1, 3, 256, 256)
pred = model(img) # (1, 1000)

Citations

@misc{tolstikhin2021mlpmixer,
    title   = {MLP-Mixer: An all-MLP Architecture for Vision},
    author  = {Ilya Tolstikhin and Neil Houlsby and Alexander Kolesnikov and Lucas Beyer and Xiaohua Zhai and Thomas Unterthiner and Jessica Yung and Daniel Keysers and Jakob Uszkoreit and Mario Lucic and Alexey Dosovitskiy},
    year    = {2021},
    eprint  = {2105.01601},
    archivePrefix = {arXiv},
    primaryClass = {cs.CV}
}

PyTorch implementation of MLP-Mixer

PyTorch implementation of MLP-Mixer MLP-Mixer: an all-MLP architecture composed of alternate token-mixing and channel-mixing operations. The token-mix

33 Nov 27, 2022

Unofficial Implementation of MLP-Mixer in TensorFlow

mlp-mixer-tf Unofficial Implementation of MLP-Mixer [abs, pdf] in TensorFlow. Note: This project may have some bugs in it. I'm still learning how to i

24 Mar 23, 2022

Implementation of Segformer, Attention + MLP neural network for segmentation, in Pytorch

Segformer - Pytorch Implementation of Segformer, Attention + MLP neural network for segmentation, in Pytorch. Install $ pip install segformer-pytorch

208 Dec 25, 2022

Implementation of "A MLP-like Architecture for Dense Prediction"

A MLP-like Architecture for Dense Prediction (arXiv) Updates (22/07/2021) Initial release. Model Zoo We provide CycleMLP models pretrained on ImageNet

244 Dec 27, 2022

🍀 Pytorch implementation of various Attention Mechanisms, MLP, Re-parameter, Convolution, which is helpful to further understand papers.⭐⭐⭐

7.7k Jan 5, 2023

PaddleViT: State-of-the-art Visual Transformer and MLP Models for PaddlePaddle 2.0+

PaddlePaddle Vision Transformers State-of-the-art Visual Transformer and MLP Models for PaddlePaddle 🤖 PaddlePaddle Visual Transformers (PaddleViT or

1k Dec 28, 2022

Keras attention models including botnet,CoaT,CoAtNet,CMT,cotnet,halonet,resnest,resnext,resnetd,volo,mlp-mixer,resmlp,gmlp,levit

Keras_cv_attention_models Keras_cv_attention_models Usage Basic Usage Layers Model surgery AotNet ResNetD ResNeXt ResNetQ BotNet VOLO ResNeSt HaloNet

319 Dec 28, 2022

Unofficial Implementation of MLP-Mixer, Image Classification Model

MLP-Mixer Unoffical Implementation of MLP-Mixer, easy to use with terminal. Train and test easly. https://arxiv.org/abs/2105.01601 MLP-Mixer is an arc

6 Dec 5, 2022

MLP-Numpy - A simple modular implementation of Multi Layer Perceptron in pure Numpy.

MLP-Numpy A simple modular implementation of Multi Layer Perceptron in pure Numpy. I used the Iris dataset from scikit-learn library for the experimen

1 Jan 1, 2022

Comments

expansion_factor on tokens is actually a bottleneck in original codebase

Thanks for your implementation. In comparing your codebase to the author's implementation, I discovered that while you have a single expansion factor in your configuration, the authors have separate values - one for tokens and one for channels.

Specifically, their channels expansion factor is 4, but their tokens expansion factor is 0.5. (The hidden_dim is the base projection size). Note that they actually use a feature count, but I'm translating to the mechanism you use in this codebase.

Thus, when executing the MixerBlock, the tokens "expansion" is actually a bottleneck.

The parameters can be verified as well in Table 1 ("Specifications of Mixer Architectures") at the top of page 4 in version 4 (the current version as of Feb 14, 2022) of their paper.

I'm not suggesting that anything necessarily needs to change in your implementation. However, if you wanted to align your codebase to be able to fully replicate the author's work, you may consider allowing for two separate parameters - token_expansion_factor and channels_expansion_factor.

Thank you again for this work, and for all your contributions generally. You are a an incredible asset to the community.

opened by chazzmoney 1
Dall-E implementation

Amazing work! How difficult is it to implement mlp into Dall-E? As the whole idea around Dall-E evolves around attention layers and transformers, I wonder if this simpler model would enable smaller, equally capable models...

opened by robvanvolt 1

An All-MLP solution for Vision, from Google AI

Related tags

Overview

MLP Mixer - Pytorch

Install

Usage

Citations

You might also like...

PyTorch implementation of MLP-Mixer

Unofficial Implementation of MLP-Mixer in TensorFlow

Implementation of Segformer, Attention + MLP neural network for segmentation, in Pytorch

Implementation of "A MLP-like Architecture for Dense Prediction"

🍀 Pytorch implementation of various Attention Mechanisms, MLP, Re-parameter, Convolution, which is helpful to further understand papers.⭐⭐⭐

PaddleViT: State-of-the-art Visual Transformer and MLP Models for PaddlePaddle 2.0+

Keras attention models including botnet,CoaT,CoAtNet,CMT,cotnet,halonet,resnest,resnext,resnetd,volo,mlp-mixer,resmlp,gmlp,levit

Unofficial Implementation of MLP-Mixer, Image Classification Model

MLP-Numpy - A simple modular implementation of Multi Layer Perceptron in pure Numpy.

Comments

expansion_factor on tokens is actually a bottleneck in original codebase

Dall-E implementation

Releases(0.1.1)

0.1.1(Feb 17, 2022)

0.1.0(Feb 17, 2022)

0.0.10(Jun 24, 2021)

0.0.9(Jun 24, 2021)

0.0.8(Jun 24, 2021)

0.0.7(May 30, 2021)

0.0.6(May 7, 2021)

0.0.5(May 7, 2021)

0.0.4(May 7, 2021)

0.0.3(May 5, 2021)

0.0.2b(May 5, 2021)

0.0.1(May 5, 2021)

Owner

Phil Wang

[NeurIPS'21] "AugMax: Adversarial Composition of Random Augmentations for Robust Training" by Haotao Wang, Chaowei Xiao, Jean Kossaifi, Zhiding Yu, Animashree Anandkumar, and Zhangyang Wang.

Negative Interactions for Improved Collaborative Filtering:

A fast implementation of bss_eval metrics for blind source separation

Implementation of Vaswani, Ashish, et al. "Attention is all you need."

PyTorch implementation of neural style transfer algorithm

Ludwig is a toolbox that allows to train and evaluate deep learning models without the need to write code.

Learning Features with Parameter-Free Layers (ICLR 2022)

SSD: Single Shot MultiBox Detector pytorch implementation focusing on simplicity

An implementation of the paper "A Neural Algorithm of Artistic Style"

[NeurIPS'20] Multiscale Deep Equilibrium Models

HMLLDB is a collection of LLDB commands to assist in the debugging of iOS apps.

ChineseBERT: Chinese Pretraining Enhanced by Glyph and Pinyin Information

[CVPR 2021] Forecasting the panoptic segmentation of future video frames

Implementation for paper LadderNet: Multi-path networks based on U-Net for medical image segmentation

The PyTorch implementation of Directed Graph Contrastive Learning (DiGCL), NeurIPS-2021

Generative Flow Networks

Quantum-enhanced transformer neural network

The codes reproduce the figures and statistics in the paper, "Controlling for multiple covariates," by Mark Tygert.

Code for Dual Contrastive Learning for Unsupervised Image-to-Image Translation, NTIRE, CVPRW 2021.

The repository contain code for building compiler using puthon.