Gauss Rank Transformation Is 100x Faster with RAPIDS and CuPy

Originally published at: Gauss Rank Transformation Is 100x Faster with RAPIDS and CuPy | NVIDIA Developer Blog

As explained in the Batch Normalization paper, training neural networks becomes way easier if its input is Gaussian. This is clear. And if your model inputs are not Gaussian, RAPIDS will just transform it to Gaussian in the blink of an eye. Gauss rank transformation is a novel standardization technique to transform input data for…