MixPro: Data Augmentation with MaskMix and Progressive Attention Labeling for Vision Transformer [Official, ICLR 2023,paper] 🔥

Qihao Zhao¹, Yangyu Huang², Wei Hu¹, Fan Zhang¹, Jun Liu³

1 Beijing University of Chemical Technology

2 Microsoft Research Asia

3 Singapore University of Technology and Design

Citation

If you find our work inspiring or use our codebase in your research, please consider giving a star ⭐️ and a citation.

@inproceedings{
zhao2023mixpro,
title={MixPro: Data Augmentation with MaskMix and Progressive Attention Labeling for Vision Transformer},
author={Qihao Zhao and Yangyu Huang and Wei Hu and Fan Zhang and Jun Liu},
booktitle={The Eleventh International Conference on Learning Representations },
year={2023},
url={https://openreview.net/forum?id=dRjWsd3gwsm} 
}

Install

We recommend using the pytorch docker nvcr>=21.05 by nvidia: https://catalog.ngc.nvidia.com/orgs/nvidia/containers/pytorch.

• Create a conda virtual environment and activate it:

conda create -n mixpro python=3.7 -y
conda activate mixpro

• Install CUDA>=10.2 with cudnn>=7 following the official installation instructions
• Install PyTorch>=1.8.0 and torchvision>=0.9.0 with CUDA>=10.2:

conda install pytorch==1.8.0 torchvision==0.9.0 cudatoolkit=10.2 -c pytorch

• Install timm==0.4.12:

pip install timm==0.4.12

• Install other requirements:

pip install opencv-python==4.4.0.46 termcolor==1.1.0 yacs==0.1.8

Data preparation

For ImageNet-1K dataset, you can download it from http://image-net.org/. We provide the following two ways to load data:

• For standard folder dataset, move validation images to labeled sub-folders. The file structure should look like:

  $ tree data
  imagenet
  ├── train
  │   ├── class1
  │   │   ├── img1.jpeg
  │   │   ├── img2.jpeg
  │   │   └── ...
  │   ├── class2
  │   │   ├── img3.jpeg
  │   │   └── ...
  │   └── ...
  └── val
      ├── class1
      │   ├── img4.jpeg
      │   ├── img5.jpeg
      │   └── ...
      ├── class2
      │   ├── img6.jpeg
      │   └── ...
      └── ...
  

• To boost the slow speed when reading images from massive small files, we also support zipped ImageNet, which includes four files:

  • train.zip, val.zip: which store the zipped folder for train and validate splits.
  • train_map.txt, val_map.txt: which store the relative path in the corresponding zip file and ground truth label. Make sure the data folder looks like this:

  $ tree data
  data
  └── ImageNet-Zip
      ├── train_map.txt
      ├── train.zip
      ├── val_map.txt
      └── val.zip
  
  $ head -n 5 data/ImageNet-Zip/val_map.txt
  ILSVRC2012_val_00000001.JPEG  65
  ILSVRC2012_val_00000002.JPEG  970
  ILSVRC2012_val_00000003.JPEG  230
  ILSVRC2012_val_00000004.JPEG  809
  ILSVRC2012_val_00000005.JPEG  516
  
  $ head -n 5 data/ImageNet-Zip/train_map.txt
  n01440764/n01440764_10026.JPEG        0
  n01440764/n01440764_10027.JPEG        0
  n01440764/n01440764_10029.JPEG        0
  n01440764/n01440764_10040.JPEG        0
  n01440764/n01440764_10042.JPEG        0

Training from scratch on ImageNet-1K

To train a MixPro with Vision Transformer on ImageNet from scratch, run:

python -m torch.distributed.launch --nproc_per_node <num-of-gpus-to-use> --master_port 12345  main.py \
--cfg <config-file> --data-path <imagenet-path> [--batch-size <batch-size-per-gpu> --output <output-directory> --tag <job-tag>]

Notes:

• To use zipped ImageNet instead of folder dataset, add --zip to the parameters.
  • To cache the dataset in the memory instead of reading from files every time, add --cache-mode part, which will shard the dataset into non-overlapping pieces for different GPUs and only load the corresponding one for each GPU.
• When GPU memory is not enough, you can try the following suggestions:
  • Use gradient accumulation by adding --accumulation-steps <steps>, set appropriate <steps> according to your need.
  • Use gradient checkpointing by adding --use-checkpoint, e.g., it saves about 60% memory when training DeiT-B. Please refer to this page for more details.
  • We recommend using multi-node with more GPUs for training very large models, a tutorial can be found in this page.
• To change config options in general, you can use --opts KEY1 VALUE1 KEY2 VALUE2, e.g., --opts TRAIN.EPOCHS 100 TRAIN.WARMUP_EPOCHS 5 will change total epochs to 100 and warm-up epochs to 5.
• For additional options, see config and run python main.py --help to get detailed message.

For example, to train MixPro DeiT Transformer with 8 GPU on a single node for 300 epochs, run:

MixPro DeiT-T:

python -m torch.distributed.launch --nproc_per_node 8 --master_port 12345  main.py \
--cfg configs/deit/deit_tiny_patch14_mask56_224_alpha1.yaml --data-path <imagenet-path> --batch-size 128

MixPro DeiT-S:

python -m torch.distributed.launch --nproc_per_node 8 --master_port 12345  main.py \
--cfg configs/deit/deit_small_patch14_mask56_224_alpha1.yaml --data-path <imagenet-path> --batch-size 128

MixPro DeiT-B:

python -m torch.distributed.launch --nproc_per_node 8 --master_port 12345  main.py \
--cfg configs/deit_base_patch14_mask112_224_alpha1_attn_all.yaml --data-path <imagenet-path> --batch-size 64 \

Config

#in config.py
#Probability of switching to mixpro when both mixup and mixpro enabled
_C.AUG.MASKMIX_PROB = 0.5
# MaskMix alpha , maskmix enabled if > 0
_C.AUG.MASKMIX_ALPHA = 1.0
# PAL 
_C.AUG.PAL_ATTN = True

Acknowledgements

The project is based on Swin and Vit(unofficial)

License

The project is released under the MIT License