SwiftFormer: Efficient Additive Attention for Transformer-based Real-time Mobile Vision Applications
Abdelrahman Shaker, Muhammad Maaz, Hanoona Rasheed, Salman Khan, Ming-Hsuan Yang, and Fahad Shahbaz Khan
- (Mar 27, 2023): Classification training and evaluation codes along with pre-trained models are released.
Comparison of our SwiftFormer Models with state-of-the-art on ImgeNet-1K. The latency is measured on iPhone 14 Neural Engine (iOS 16).
Comparison with different self-attention modules. (a) is a typical self-attention. (b) is the transpose self-attention, where the self-attention operation is applied across channel feature dimensions (d×d) instead of the spatial dimension (n×n). (c) is the separable self-attention of MobileViT-v2, it uses element-wise operations to compute the context vector from the interactions of Q and K matrices. Then, the context vector is multiplied by V matrix to produce the final output. (d) Our proposed efficient additive self-attention. Here, the query matrix is multiplied by learnable weights and pooled to produce global queries. Then, the matrix K is element-wise multiplied by the broadcasted global queries, resulting the global context representation.
Abstract
Self-attention has become a defacto choice for capturing global context in various vision applications. However, its quadratic computational complexity with respect to image resolution limits its use in real-time applications, especially for deployment on resource-constrained mobile devices. Although hybrid approaches have been proposed to combine the advantages of convolutions and self-attention for a better speed-accuracy trade-off, the expensive matrix multiplication operations in self-attention remain a bottleneck. In this work, we introduce a novel efficient additive attention mechanism that effectively replaces the quadratic matrix multiplication operations with linear element-wise multiplications. Our design shows that the key-value interaction can be replaced with a linear layer without sacrificing any accuracy. Unlike previous state-of-the-art methods, our efficient formulation of self-attention enables its usage at all stages of the network. Using our proposed efficient additive attention, we build a series of models called "SwiftFormer" which achieves state-of-the-art performance in terms of both accuracy and mobile inference speed. Our small variant achieves 78.5% top-1 ImageNet-1K accuracy with only 0.8~ms latency on iPhone 14, which is more accurate and 2x faster compared to MobileViT-v2.Model | Top-1 accuracy | #params | GMACs | Latency | Ckpt | CoreML |
---|---|---|---|---|---|---|
SwiftFormer-XS | 75.7% | 3.5M | 0.4G | 0.7ms | XS | XS |
SwiftFormer-S | 78.5% | 6.1M | 1.0G | 0.8ms | S | S |
SwiftFormer-L1 | 80.9% | 12.1M | 1.6G | 1.1ms | L1 | L1 |
SwiftFormer-L3 | 83.0% | 28.5M | 4.0G | 1.9ms | L3 | L3 |
The latency reported in SwiftFormer for iPhone 14 (iOS 16) uses the benchmark tool from XCode 14.
conda
virtual environment is recommended.
conda create --name=swiftformer python=3.9
conda activate swiftformer
pip install torch==1.11.0+cu113 torchvision==0.12.0+cu113 --extra-index-url https://download.pytorch.org/whl/cu113
pip install timm
Download and extract ImageNet train and val images from http://image-net.org. The training and validation data are expected to be in the train
folder and val
folder respectively:
|-- /path/to/imagenet/
|-- train
|-- val
We provide training script for all models in dist_train.sh
using PyTorch distributed data parallel (DDP).
To train SwiftFormer models on an 8-GPU machine:
sh dist_train.sh /path/to/imagenet 8
Note: specify which model command you want to run in the script. To reproduce the results of the paper, use 16-GPU machine with batch-size of 128 or 8-GPU machine with batch size of 256. Auto Augmentation, CutMix, MixUp are disabled for SwiftFormer-XS only.
On a Slurm-managed cluster, multi-node training can be launched as
sbatch slurm_train.sh /path/to/imagenet SwiftFormer_XS
Note: specify slurm specific paramters in slurm_train.sh
script.
We provide an example test script dist_test.sh
using PyTorch distributed data parallel (DDP).
For example, to test SwiftFormer-XS on an 8-GPU machine:
sh dist_test.sh SwiftFormer_XS 8 weights/SwiftFormer_XS_ckpt.pth
if you use our work, please consider citing us:
@article{Shaker2023SwiftFormer,
title={SwiftFormer: Efficient Additive Attention for Transformer-based Real-time Mobile Vision Applications},
author={Shaker, Abdelrahman and Maaz, Muhammad and Rasheed, Hanoona and Khan, Salman and Yang, Ming-Hsuan and Khan, Fahad Shahbaz},
journal={arXiv preprint arXiv:2303.15446},
year={2023}
}
If you have any question, please create an issue on this repository or contact at abdelrahman.youssief@mbzuai.ac.ae.
Our code base is based on LeViT and EfficientFormer repositories. We thank authors for their open-source implementation.