BEiT

BEiT: BERT Pre-Training of Image Transformers, arxiv

PaddlePaddle training/validation code and pretrained models for BEiT.

The official and 3rd party pytorch implementation are here.

This implementation is developed by PaddleViT.

BEiT Model Overview

Update

• Update (2022-03-24): Code is refactored and bugs are fixed.
• Update (2021-10-19): Bug fix and weights links are updated.
• Update (2021-09-27): Code is released and ported weights are uploaded.

Models Zoo

Model	Acc@1	Acc@5	#Params	FLOPs	Image Size	Crop_pct	Interpolation	Link
beit_base_patch16_224	85.21	97.66	87M	12.7G	224	0.9	bicubic	google/baidu
beit_base_patch16_384	86.81	98.14	87M	37.3G	384	1.0	bicubic	google/baidu
beit_large_patch16_224	87.48	98.30	304M	45.0G	224	0.9	bicubic	google/baidu
beit_large_patch16_384	88.40	98.60	304M	131.7G	384	1.0	bicubic	google/baidu
beit_large_patch16_512	88.60	98.66	304M	234.0G	512	1.0	bicubic	google/baidu

Note:

The results are evaluated on ImageNet2012 validation set.

These models have been fine-tuned (ImageNet 22k -> 1k), weights are ported from here

Note : We are developing the pretraining using DALL-E, which is not supported right now.

Data Preparation

ImageNet2012 dataset is used in the following file structure:

│imagenet/
├──train_list.txt
├──val_list.txt
├──train/
│  ├── n01440764
│  │   ├── n01440764_10026.JPEG
│  │   ├── n01440764_10027.JPEG
│  │   ├── ......
│  ├── ......
├──val/
│  ├── n01440764
│  │   ├── ILSVRC2012_val_00000293.JPEG
│  │   ├── ILSVRC2012_val_00002138.JPEG
│  │   ├── ......
│  ├── ......

• train_list.txt: list of relative paths and labels of training images. You can download it from: google/baidu
• val_list.txt: list of relative paths and labels of validation images. You can download it from: google/baidu

Usage

To use the model with pretrained weights, download the .pdparam weight file and change related file paths in the following python scripts. The model config files are located in ./configs/.

For example, assume weight file is downloaded in ./beit_base_patch16_224.pdparams, to use the beit_base_patch16_224 model in python:

from config import get_config
from beit import build_beit as build_model
# config files in ./configs/
config = get_config('./configs/beit_base_patch16_224.yaml')
# build model
model = build_model(config)
# load pretrained weights
model_state_dict = paddle.load('./beit_base_patch16_224.pdparams')
model.set_state_dict(model_state_dict)

Evaluation

To evaluate model performance on ImageNet2012, run the following script using command line:

sh run_eval_multi.sh

or

CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \
python main_multi_gpu.py \
-cfg='./configs/beit_base_patch16_224.yaml' \
-dataset='imagenet2012' \
-batch_size=256 \
-data_path='/dataset/imagenet' \
-eval \
-pretrained='./beit_base_patch16_224.pdparams' \
-amp

Note: if you have only 1 GPU, change device number to CUDA_VISIBLE_DEVICES=0 would run the evaluation on single GPU.

Training

To train the model on ImageNet2012, run the following script using command line:

sh run_train_multi.sh

or

CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \
python main_multi_gpu.py \
-cfg='./configs/beit_base_patch16_224.yaml' \
-dataset='imagenet2012' \
-batch_size=256 \
-data_path='/dataset/imagenet' \
-amp

Note: it is highly recommanded to run the training using multiple GPUs / multi-node GPUs.

Reference

@article{beit,
      title={{BEiT}: {BERT} Pre-Training of Image Transformers}, 
      author={Hangbo Bao and Li Dong and Furu Wei},
      year={2021},
      eprint={2106.08254},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}