Beta Version 1.6.1

Added

  • None

Fixes

  • None

Changes

  • The layer matrices will be only generated upon training for neural networks. Previously, it was generated upon calling createLayers() and addLayer() functions.

  • Made neural network calculations more efficient by concatenating bias matrix to feature matrix when calling train() and predict() functions. Previously it was concatenated when calling the forward propagate function.


This site uses Just the Docs, a documentation theme for Jekyll.