SimpleNN

SimpleNN is a simple package used for training Convolutional Neural Network (CNN) with following supports:

• MATLAB/Octave package
• Python (implemented in Tensorflow)

Currently, both implementations support two optimization methods: Newton method and stochastic gradient method (SG). The implementation document of Newton method is available at https://www.csie.ntu.edu.tw/~cjlin/papers/cnn/newton-CNN.pdf.

Authors:

• MATLAB/Octave: CHIEN-CHIH WANG, KENT LOONG TAN, CHIH-JEN LIN
• Python: PENGRUI QUAN, CHIH-JEN LIN