plotStepFunction

number of steps on interval (0, 1)

nSteps

smoothing parameter. Small values results in high smoothness.

smoothSteps

logical indicating if the derivative of the step function should be plotted.

derivative

further arguments to be passed to plot

Change parameters to see how the shape of the step function is affected.
Note that setting smoothSteps very high results in a large deriative,
this might cause instability during training.

Training of general classification and regression neural networks using
gradient descent. Special features include a function for training replicator
neural networks and a function for training autoencoders. Multiple activation and
cost functions (including Huber and pseudo-Huber) are supported, as well as L1
and L2 regularization, momentum, early stopping and the possibility to specify
a learning rate schedule. The package contains a vectorized gradient descent
implementation which facilitates faster training through batch learning.

plotStepFunction: Plot the step function used in `replicator`

Description

Usage

Arguments