layer_gaussian_dropout

As it is a regularization layer, it is only active at training time.

Interface to 'Keras' <https://keras.io>, a high-level neural
networks 'API'. 'Keras' was developed with a focus on enabling fast experimentation,
supports both convolution based networks and recurrent networks (as well as
combinations of the two), and runs seamlessly on both 'CPU' and 'GPU' devices.

Tomasz Kalinowski

keras

R Interface to 'Keras'

Daniel Falbel

JJ Allaire

François Chollet

RStudio 

Google 

Yuan Tang

Wouter Van Der Bijl

Martin Studer

Sigrid Keydana

layer_gaussian_dropout function

<dl><dt>object</dt>
<dd>What to compose the new <code>Layer</code> instance with. Typically a
Sequential model or a Tensor (e.g., as returned by <code>layer_input()</code>).
The return value depends on <code>object</code>. If <code>object</code> is:<ul>
<li>missing or <code>NULL</code>, the <code>Layer</code> instance is returned.</li>
<li>a <code>Sequential</code> model, the model with an additional layer is returned.</li>
<li>a Tensor, the output tensor from <code>layer_instance(object)</code> is returned.</li>
</ul></dd>
<dt>rate</dt>
<dd>float, drop probability (as with <code>Dropout</code>). The multiplicative
noise will have standard deviation <code>sqrt(rate / (1 - rate))</code>.</dd>
<dt>seed</dt>
<dd>Integer, optional random seed to enable deterministic behavior.</dd>
<dt>...</dt>
<dd>standard layer arguments.</dd></dl>

Arguments

Arbitrary. Use the keyword argument <code>input_shape</code> (list
of integers, does not include the samples axis) when using this layer as
the first layer in a model.

Input shape

Output shape

Apply multiplicative 1-centered Gaussian noise. — layer_gaussian_dropout

<dl>

<dt>object</dt>
<dd>What to compose the new <code>Layer</code> instance with. Typically a
Sequential model or a Tensor (e.g., as returned by <code>layer_input()</code>).
The return value depends on <code>object</code>. If <code>object</code> is:<ul>
<li>missing or <code>NULL</code>, the <code>Layer</code> instance is returned.</li>
<li>a <code>Sequential</code> model, the model with an additional layer is returned.</li>
<li>a Tensor, the output tensor from <code>layer_instance(object)</code> is returned.</li>
</ul></dd>


<dt>rate</dt>
<dd>float, drop probability (as with <code>Dropout</code>). The multiplicative
noise will have standard deviation <code>sqrt(rate / (1 - rate))</code>.</dd>


<dt>seed</dt>
<dd>Integer, optional random seed to enable deterministic behavior.</dd>


<dt>...</dt>
<dd>standard layer arguments.</dd>

</dl>

layer_gaussian_dropout: Apply multiplicative 1-centered Gaussian noise.

Description

Usage

Arguments

Input shape

Output shape

References

See Also