On the Importance of Normalisation Layers in Deep Learning with Piecewise Linear Activation Units

Benchmark Model Rank Results
image-classification-on-cifar-10MIM–Percentage correct: 91.5
image-classification-on-cifar-100MIM–Percentage correct: 70.8
image-classification-on-mnistMIM–Percentage error: 0.4
image-classification-on-svhnMIM–Percentage error: 2.0