loss

2d tensor with shape (batch size, dimension of input dataset).

<code><a rd-options="" href="/link/character?package=gnn&version=0.0-2" data-mini-rdoc="gnn::character">character</a></code> string indicating the type of
 loss used. Currently available are the mean
 squared error (<code>"MSE"</code>), binary cross entropy
 (<code>"binary.cross"</code>)
 and (kernel) maximum mean discrepancy (<code>"MMD"</code>).

type

additional arguments passed to the underlying loss function;
 at the moment, this is only affects <code>type = "MMD"</code> for which
 <code>"bandwidth"</code> can be provided.

&#8230;

Implementation of various loss functions to measure
 statistical discrepancy between two datasets.

univar

Tools to set up, train, store, load, investigate and analyze
generative neural networks. In particular, functionality for
generative moment matching networks is provided.
See Hofert, Prasad, Zhu (2018) <arXiv:1811.00683>
for more details.

Marius Hofert

Generative Neural Networks

Avinash Prasad

loss function

<code><a rd-options='' href='character'>character</a></code> string indicating the type of
 loss used. Currently available are the mean
 squared error (<code>"MSE"</code>), binary cross entropy
 (<code>"binary.cross"</code>)
 and (kernel) maximum mean discrepancy (<code>"MMD"</code>).

loss: Loss Function

Description

Usage

Arguments

Value

References

See Also