MNIST learning rate

[glangford]

The paper says in 4:

Our implementation is in TensorFlow (Abadi et al. [2016]) and we use the Adam optimizer (Kingma and Ba [2014]) with its TensorFlow default parameters, including the exponentially decaying learning rate

The TensorFlow defaults for Adam are described here:
https://www.tensorflow.org/api_docs/python/tf/train/AdamOptimizer

The current capsulenet.py uses lr_decay as a callback to modify the learning rate, but there isn't any evidence that the paper follows this method. Should the lr_decay callback be removed since Adam already decays the learning rate?
(update: the TensorFlow and Keras defaults for Adam appear to be the same)

[XifengGuo]

@glangford As I said in README.md, I'm not sure if the paper used this learning rate decay method. I found that adopting lr_decay can lead to faster convergence. You can remove it and train for more epochs, it's your choice.