NN_implementation

✏ Multi-Layer Neural Network Implementation

• Line-by-line implementation of multi-layer feed forward neural networks with back propagation.

• All code is based on "Neural Networks and Deep Learning" by Michael Nielsen.

Version

• NN_v1 : MLP implemented by Python & Numpy. Sigmoid, MSE, No opt. 90.1 ~ 91.3% Accuracy

• NN_v2 : Improved performance from NN_v1. ReLU, CrossEntropy, He, L2, Data Augmentation. 97.2 ~ 97.5% Accuracy

Update Note

• 21-08-16 : add NN.v1

• 21-08-18 : add NN.v2

References.

[1] "Neural Networks and Deep Learning"