compile(optimizer=’rmsprop’)

Net.compile(optimizer='rmsprop', options=None, random_state=None)

Compile the neural network with Root Mean Square Propagation (RMSprop) optimizer.

See also

For documentation for the rest of the parameters, see Net.compile

Options:

learning_ratefloat, default 0.001

The learning rate used to update the weights during training.

rhofloat, default 0.9

The decay rate for the running average of the squared gradients.