MLP-Mixer, ResMLP and gMLP

Category: MLP · Code: timm/models/mlp_mixer.py · Architectures: 15 · Pretrained weights: 19 (14 with ImageNet-1k results)

All-MLP architectures that alternate token mixing across patches with channel mixing: MLP-Mixer, ResMLP and gMLP (gating with a spatial projection). The module docstring notes the gMLP implementation is not verified.

Papers

• MLP-Mixer: An all-MLP Architecture for Vision
• ResMLP: Feedforward networks for image classification with data-efficient training
• Pay Attention to MLPs
• ImageNet-21K Pretraining for the Masses
• Emerging Properties in Self-Supervised Vision Transformers

Usage

See the Quickstart for loading pretrained weights, image preprocessing and fine-tuning, and Feature Extraction for using the models as backbones.

>>> import timm
>>> model = timm.create_model('gmixer_24_224.ra3_in1k', pretrained=True)
>>> data_config = timm.data.resolve_model_data_config(model)

Model API support

✓ supported, ✓ (square) works for square sizes only, ✗ not supported, – not needed (the model already accepts any input size). Checked by running each feature on a randomly initialized model (MlpMixer with gmlp_ti16_224 at 224), other architectures of the same class are expected to behave the same. See the Model API page for details.

Architectures

GMACs and parameter counts are for the default input size and classifier of each architecture, from the benchmark results.

Pretrained weights

Top-1 is ImageNet-1k validation accuracy from results-imagenet.csv at the listed eval size, - if the weight has no ImageNet-1k result (e.g. pretraining or image-text weights). Classes is the size of the pretrained classifier, none for weights without one.

Weights are grouped by source: the paper or origin given in the pretrained config, else a Hub organization other than timm, else the first part of the pretrained tag (architecture.tag). Sources with a single weight are listed under other.

fb tags

goog tags

miil tags

ra3 tags

Citation

@article{tolstikhin2021,
  title={MLP-Mixer: An all-MLP Architecture for Vision},
  author={Tolstikhin, Ilya and Houlsby, Neil and Kolesnikov, Alexander and Beyer, Lucas and Zhai, Xiaohua and Unterthiner,
        Thomas and Yung, Jessica and Keysers, Daniel and Uszkoreit, Jakob and Lucic, Mario and Dosovitskiy, Alexey},
  journal={arXiv preprint arXiv:2105.01601},
  year={2021}
}

@misc{touvron2021resmlp,
      title={ResMLP: Feedforward networks for image classification with data-efficient training},
      author={Hugo Touvron and Piotr Bojanowski and Mathilde Caron and Matthieu Cord and Alaaeldin El-Nouby and
        Edouard Grave and Armand Joulin and Gabriel Synnaeve and Jakob Verbeek and Hervé Jégou},
      year={2021},
      eprint={2105.03404},
}

@misc{liu2021pay,
      title={Pay Attention to MLPs},
      author={Hanxiao Liu and Zihang Dai and David R. So and Quoc V. Le},
      year={2021},
      eprint={2105.08050},
}