An implementation of batch normalization with early stopping
for a simple feedforward neural net with 2 hidden layers on the MNIST dataset.
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
by Sergey Ioffe and Christian Szegedy
Paper url: https://arxiv.org/pdf/1502.03167.pdf
Thanks to R2RT for their implementation in
http://r2rt.com/implementing-batch-normalization-in-tensorflow.html
If it wasn't for that I would probably still be struggling with tensorflow dependencies...
Use and abuse to your heart's content, just give credit ;)
Do let me know if you notice something wrong with the code.
TensorFlow version: 1.0.1
Python version 3.x
Dimitris