xwgfree / deep-learning-with-coherent-nanophotonic-circuits Goto Github PK

Simulation of the Nanophotonic Network in paper "Deep learning with coherent nanophotonic circuits" [1]

Vowel Classification dataset provided by [2]

ONN.py: Implementation of the nonlinear saturable absorber activation function with backpropgation algorithm according to the steps in the supplementary information of [1]

deterding_pytorch.py: Implementation of a similar network in Pytorch with ReLU() acitivation function for comparison purposes.

train.csv: vowel samples for training the network. See [2] for description test.csv: vowel samples for testing trained network. See [2] for description

References:

[1]Shen, Y., Harris, N., Skirlo, S. et al. Deep learning with coherent nanophotonic circuits. Nature Photon 11, 441–446 (2017). https://doi.org/10.1038/nphoton.2017.93

[2]Deterding Dataset https://archive.ics.uci.edu/ml/datasets/Connectionist+Bench+(Vowel+Recognition+-+Deterding+Data)