Supervised Learning Based on Temporal Coding in Spiking Neural Networks

This project is spiking neural networks for XOR and MNIST using PyTorch.
Look at this article

XOR networks architecture: DNN(2, 4, 2)

Epoch: 90   loss: 0.00000000    pred: 1 1 0 0   label: 1 1 0 0  4 / 4
Epoch: 91   loss: 0.00000000    pred: 1 1 0 0   label: 1 1 0 0  4 / 4
Epoch: 92   loss: 0.00000000    pred: 1 1 0 0   label: 1 1 0 0  4 / 4
Epoch: 93   loss: 0.00000000    pred: 1 1 0 0   label: 1 1 0 0  4 / 4
Epoch: 94   loss: 0.00000000    pred: 1 1 0 0   label: 1 1 0 0  4 / 4
Epoch: 95   loss: 0.00000000    pred: 1 1 0 0   label: 1 1 0 0  4 / 4
Epoch: 96   loss: 0.00000000    pred: 1 1 0 0   label: 1 1 0 0  4 / 4
Epoch: 97   loss: 0.00000000    pred: 1 1 0 0   label: 1 1 0 0  4 / 4
Epoch: 98   loss: 0.00000000    pred: 1 1 0 0   label: 1 1 0 0  4 / 4
Epoch: 99   loss: 0.00000000    pred: 1 1 0 0   label: 1 1 0 0  4 / 4

MNIST networks architecture: DNN(784, 800, 10)

not yet

Requirements

• Python 3.6
• PyTorch 1.1.0
• CUDA>=9.0

Authors

• HanSeokhyeon - Initial work - HanSeokhyeon

See also the list of contributors who participated in this project.

License

This project is licensed under the MIT License - see the LICENSE.md file for details