Skip to content

Latest commit

 

History

History
100 lines (61 loc) · 3.49 KB

README.md

File metadata and controls

100 lines (61 loc) · 3.49 KB
Raw

Nested Hierarchical Transformer Official Jax Implementation

NesT is a simple method, which aggregates nested local transformers on image blocks. The idea makes vision transformers attain better accuracy, data efficiency, and convergence on the ImageNet benchmark. NesT can be scaled to small datasets to match convnet accuracy.

This is not an officially supported Google product.

Pretrained Models and Results

Model Accuracy Checkpoint path
Nest-B 83.8 gs://gresearch/nest-checkpoints/nest-b_imagenet
Nest-S 83.3 gs://gresearch/nest-checkpoints/nest-s_imagenet
Nest-T 81.5 gs://gresearch/nest-checkpoints/nest-t_imagenet

Note: Accuracy is evaluated on the ImageNet2012 validation set.

Tensorbord.dev

See ImageNet training logs at Tensorboard.dev.

Colab

Colab is available for test: https://colab.sandbox.google.com/github/google-research/nested-transformer/blob/main/colab.ipynb

Pytorch re-implementation

The timm library has incorporated NesT and pre-trained models in Pytorch.

Instruction on Image Classification

Environment setup

virtualenv -p python3 --system-site-packages nestenv
source nestenv/bin/activate

pip install -r requirements.txt

Evaluate on ImageNet

At the first time, download ImageNet following tensorflow_datasets instruction from command lines. Optionally, download all pre-trained checkpoints

bash ./checkpoints/download_checkpoints.sh

Run the evaluation script to evaluate NesT-B.

python main.py --config configs/imagenet_nest.py --config.eval_only=True \
  --config.init_checkpoint="./checkpoints/nest-b_imagenet/ckpt.39" \
  --workdir="./checkpoints/nest-t_imagenet_eval"

Train on ImageNet

The default configuration trains NesT-B on TPUv2 8x8 with per device batch size 16.

python main.py --config configs/imagenet_nest.py --jax_backend_target=<TPU_IP_ADDRESS> --jax_xla_backend="tpu_driver" --workdir="./checkpoints/nest-b_imagenet"

Note: See jax/cloud_tpu_colab for info about TPU_IP_ADDRESS.

Train NesT-T on 8 GPUs.

python main.py --config configs/imagenet_nest_tiny.py --workdir="./checkpoints/nest-t_imagenet_8gpu"

The codebase does not support multi-node GPU training (>8 GPUs). The models reported in our paper is trained using TPU with 1024 total batch size.

Train on CIFAR

# Recommend to train on 2 GPUs. Training NesT-T can use 1 GPU.
CUDA_VISIBLE_DEVICES=0,1 python  main.py --config configs/cifar_nest.py --workdir="./checkpoints/nest_cifar"

Cite

@inproceedings{zhang2021aggregating,
  title={Nested Hierarchical Transformer: Towards Accurate, Data-Efficient and Interpretable Visual Understanding},
  author={Zizhao Zhang and Han Zhang and Long Zhao and Ting Chen and and Sercan Ö. Arık and Tomas Pfister},
  booktitle={AAAI Conference on Artificial Intelligence (AAAI)},
  year={2022}
}