Skip to content

Latest commit

 

History

History
 
 

CycleMLP

README.md

CycleMLP: A MLP-like Architecture for Dense Prediction, arXiv

PaddlePaddle training/validation code and pretrained models for CycleMLP.

The official and 3rd party pytorch implementation are here.

This implementation is developed by PPViT.

drawing

CycleMLP Model Overview

Update

Update (2021-09-24): Code is released and ported weights are uploaded.

Models Zoo

Model Acc@1 Acc@5 #Params Image Size Crop_pct Interpolation Link
cyclemlp_b1 78.85 94.60 15.1M 224 0.9 bicubic google/baidu(mnbr)
cyclemlp_b2 81.58 95.81 26.8M 224 0.9 bicubic google/baidu(jwj9)
cyclemlp_b3 82.42 96.07 38.3M 224 0.9 bicubic google/baidu(v2fy)
cyclemlp_b4 82.96 96.33 51.8M 224 0.875 bicubic google/baidu(fnqd)
cyclemlp_b5 83.25 96.44 75.7M 224 0.875 bicubic google/baidu(s55c)

*The results are evaluated on ImageNet2012 validation set.

Note: CycleMLP weights are ported from here

Notebooks

We provide a few notebooks in aistudio to help you get started:

*(coming soon)*

Requirements

Data

ImageNet2012 dataset is used in the following folder structure:

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

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 the downloaded weight file is stored in ./cyclemlp_b1.pdparams, to use the cyclemlp_b1 model in python:

from config import get_config
from cyclemlp import build_cyclemlp as build_model
# config files in ./configs/
config = get_config('./configs/cyclemlp_b1.yaml')
# build model
model = build_model(config)
# load pretrained weights, .pdparams is NOT needed
model_state_dict = paddle.load('./cyclemlp_b1')
model.set_dict(model_state_dict)

Evaluation

To evaluate CycleMLP model performance on ImageNet2012 with a single GPU, run the following script using command line:

sh run_eval.sh

or

CUDA_VISIBLE_DEVICES=0 \
python main_single_gpu.py \
    -cfg='./configs/cyclemlp_b1.yaml' \
    -dataset='imagenet2012' \
    -batch_size=16 \
    -data_path='/dataset/imagenet' \
    -eval \
    -pretrained='./cyclemlp_b1'
Run evaluation using multi-GPUs: 
sh run_eval_multi.sh

or

CUDA_VISIBLE_DEVICES=0,1,2,3 \
python main_multi_gpu.py \
    -cfg='./configs/cyclemlp_b1.yaml' \
    -dataset='imagenet2012' \
    -batch_size=16 \
    -data_path='/dataset/imagenet' \
    -eval \
    -pretrained='./cyclemlp_b1'

Training

To train the CycleMLP model on ImageNet2012 with single GPUs, run the following script using command line:

sh run_train.sh

or

CUDA_VISIBLE_DEVICES=0 \
python main_single_gpu.py \
  -cfg='./configs/cyclemlp_b1.yaml' \
  -dataset='imagenet2012' \
  -batch_size=32 \
  -data_path='/dataset/imagenet' \
Run training using multi-GPUs: 
sh run_train_multi.sh

or

CUDA_VISIBLE_DEVICES=0,1,2,3 \
python main_multi_gpu.py \
    -cfg='./configs/cyclemlp_b1.yaml' \
    -dataset='imagenet2012' \
    -batch_size=16 \
    -data_path='/dataset/imagenet' \

Visualization Attention Map

(coming soon)

Reference

@article{chen2021cyclemlp,
  title={CycleMLP: A MLP-like Architecture for Dense Prediction},
  author={Chen, Shoufa and Xie, Enze and Ge, Chongjian and Liang, Ding and Luo, Ping},
  journal={arXiv preprint arXiv:2107.10224},
  year={2021}
}