Hi, is SpatialDropout2D in Keras possible in MxNet?
Hi. I want to know about an example that how we can know about a linear model is not interested in the interact between features. Please.
and why don’t we need to normalize with a trained model and a test sample, we will not remove any nodes (thus no normalization).