/Adaptively-Connected-Neural-Networks

A re-implementation of our CVPR 2019 paper "Adaptively Connected Neural Networks"

Primary LanguagePython

Adaptively Connected Neural Networks

A re-implementation of our CVPR 2019 paper "Adaptively Connected Neural Networks" (https://arxiv.org/abs/1904.03579).

Guangrun Wang , Keze Wang, Liang Lin

Sun Yat-sen University (SYSU)

intro

Table of Contents

  1. Introduction
  2. ImageNet
  3. Cora
  4. Citation

Introduction

This repository contains the training & testing code of Adaptively Connected Neural Networks (ACNet) on ImageNet and Cora.

ImageNet

  • Training and testing curve on ImageNet:

    curves

  • ImageNet accuracy and pretrained model (baidu pan code: ow9z):

Model Top 5 Error Top 1 Error Download
ResNet50 6.9% 23.6% ⬇️
ResNet50-ACNet 6.4% 22.5% google drive: ⬇️ and baidu pan: ⬇️
ResNet50-ACNet-pixel-aware 6.4% 22.5% google drive: ⬇️ and baidu pan: ⬇️
  • Training script:
cd cnn/dataset-aware/ or cd cnn/pixel-aware/

python imagenet-resnet.py  --gpu 0,1,2,3,4,5,6,7   --data [ROOT-OF-IMAGENET-DATASET]  --log_dir  [ROOT-OF-TRAINING-LOG-AND-MODEL] 
  • Testing script:
cd cnn/dataset-aware/ or cd cnn/pixel-aware/

python imagenet-resnet.py  --gpu 0,1,2,3,4,5,6,7   --data [ROOT-OF-IMAGENET-DATASET]  --log_dir  [ROOT-OF-TEST-LOG] --load   [ROOT-TO-LOAD-MODEL]  --eval

Cora

Coming soon ...

Citation

If you use these models in your research, please cite:

@inproceedings{wang2019adaptively,

  title={Adaptively Connected Neural Networks},

  author={Wang, Guangrun and Wang, Keze and Lin, Liang},

  booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},

  year={2019}}

Dependencies

  • Python 2.7 or 3
  • TensorFlow >= 1.3.0
  • Tensorpack The code depends on Yuxin Wu's Tensorpack. For convenience, we provide a stable version 'tensorpack-installed' in this repository.
    # install tensorpack locally:
    cd tensorpack-installed
    python setup.py install --user