tf.keras and TensorFlow: Batch Normalization to train deep neural networks faster | by Chris Rawles | Towards Data Science
François Chollet on Twitter: "10) Some layers, in particular the `BatchNormalization` layer and the `Dropout` layer, have different behaviors during training and inference. For such layers, it is standard practice to expose
tf.keras and TensorFlow: Batch Normalization to train deep neural networks faster | by Chris Rawles | Towards Data Science