dzy-cxy / progressive_tandem_learning

The official release of the progressive tandem learning framework.

Progressive Tandem Learning Framework

The official code release for the progressive tandem learning (PTL) framework.

Requirements

The code has been tested with python=3.8, pytorch=1.9.0, torchvision=0.10.0

Usage

This repo contains the code for image classification on the Cifar10 dataset with the AlexNet and VGG11 architectures.

• Step 1: Train baseline ANN models

$ python cifar10_AlexNet_ann.py
$ python cifar10_vgg11_ann.py

• Step 2: Train SNN models with the PTL framework

$ python cifar10_AlexNet_snn.py
$ python cifar10_vgg11_snn.py

Citation

If you find this code useful, please consider citing the following paper.

@article{wu2020progressive,
  title={Progressive tandem learning for pattern recognition with deep spiking neural networks},
  author={Wu, Jibin and Xu, Chenglin and Han, Xiao and Zhou, Daquan and Zhang, Malu and Li, Haizhou and Tan, Kay Chen},
  journal={IEEE transactions on pattern analysis and machine intelligence},
  year={2021}
}

Contact

For queries please contact [e-mail]