activation.hpp File Reference

Activation functions. More...

#include <iostream>
#include <utility>
#include <core/core.hpp>
#include <utils/fqn.hpp>

Go to the source code of this file.

Classes
class	iganet::ActivationFunction
	Abstract activation function structure. More...
class	iganet::BatchNorm
	Batch Normalization as described in the paper. More...
class	iganet::CELU
	Continuously Differentiable Exponential Linear Units activation function. More...
class	iganet::ELU
	Exponential Linear Units activation function. More...
class	iganet::GELU
	Gaussian Error Linear Units activation function. More...
class	iganet::GLU
	Grated Linear Units activation function. More...
class	iganet::GroupNorm
	Group Normalization over a mini-batch of inputs as described in the paper Group Normalization, https://arxiv.org/abs/1803.08494. More...
class	iganet::GumbelSoftmax
	Gumbel-Softmax distribution activation function. More...
class	iganet::Hardshrink
	Hard shrinkish activation function. More...
class	iganet::Hardsigmoid
	Hardsigmoid activation function. More...
class	iganet::Hardswish
	Hardswish activation function. More...
class	iganet::Hardtanh
	Hardtanh activation function. More...
class	iganet::InstanceNorm
	Instance Normalization as described in the paper. More...
class	iganet::LayerNorm
	Layer Normalization as described in the paper. More...
class	iganet::LeakyReLU
	Leaky ReLU activation function. More...
class	iganet::LocalResponseNorm
	Local response Normalization. More...
class	iganet::LogSigmoid
	LogSigmoid activation function. More...
class	iganet::LogSoftmax
	LogSoftmax activation function. More...
class	iganet::Mish
	Mish activation function. More...
class	iganet::None
	No-op activation function. More...
class	iganet::Normalize
	Lp Normalization. More...
class	iganet::PReLU
	PReLU activation function. More...
class	iganet::ReLU
	ReLU activation function. More...
class	iganet::ReLU6
	ReLU6 activation function. More...
class	iganet::RReLU
	Randomized ReLU activation function. More...
class	iganet::SELU
	SELU activation function. More...
class	iganet::Sigmoid
	Sigmoid activation function. More...
class	iganet::SiLU
	Sigmoid Linear Unit activation function. More...
class	iganet::Softmax
	Softmax activation function. More...
class	iganet::Softmin
	Softmin activation function. More...
class	iganet::Softplus
	Softplus activation function. More...
class	iganet::Softshrink
	Softshrink activation function. More...
class	iganet::Softsign
	Softsign activation function. More...
class	iganet::Tanh
	Tanh activation function. More...
class	iganet::Tanhshrink
	Tanhshrink activation function. More...
class	iganet::Threshold
	Threshold activation function. More...

Namespaces
namespace	iganet

Enumerations
enum class	iganet::activation : short_t {   iganet::none = 0 , iganet::batch_norm = 1 , iganet::celu = 2 , iganet::elu = 3 ,   iganet::gelu = 4 , iganet::glu = 5 , iganet::group_norm = 6 , iganet::gumbel_softmax = 7 ,   iganet::hardshrink = 9 , iganet::hardsigmoid = 8 , iganet::hardswish = 10 , iganet::hardtanh = 11 ,   iganet::instance_norm = 12 , iganet::layer_norm = 13 , iganet::leaky_relu = 14 , iganet::local_response_norm = 15 ,   iganet::logsigmoid = 16 , iganet::logsoftmax = 17 , iganet::mish = 18 , iganet::normalize = 19 ,   iganet::prelu = 20 , iganet::relu = 21 , iganet::relu6 = 22 , iganet::rrelu = 23 ,   iganet::selu = 24 , iganet::sigmoid = 25 , iganet::silu = 26 , iganet::softmax = 27 ,   iganet::softmin = 28 , iganet::softplus = 29 , iganet::softshrink = 30 , iganet::softsign = 31 ,   iganet::tanh = 32 , iganet::tanhshrink = 33 , iganet::threshold = 34 }
	Enumerator for nonlinear activation functions. More...

Functions
std::ostream &	iganet::operator<< (std::ostream &os, const ActivationFunction &obj)
	Print (as string) an ActivationFunction object.

Detailed Description

Activation functions.

Author

Matthias Moller

Copyright

This file is part of the IgANet project

This Source Code Form is subject to the terms of the Mozilla Public License, v. 2.0. If a copy of the MPL was not distributed with this file, You can obtain one at http://mozilla.org/MPL/2.0/.