pytorch::MaxUnpool2d Class Reference

#include <Pooling.hpp>

Inheritance diagram for pytorch::MaxUnpool2d:

Public Member Functions
	MaxUnpool2d (const pooling_params_t &params, const MaxPool2d *mp_ref)
	Constructs the MaxUnpool2d layer. Requires pooling parameters that are functionally equivalent to the convolutional parameters. More...
std::vector< tensor >	forward (const std::vector< tensor > &input)
	Implements the forward pass. More...
std::vector< tensor >	operator() (const std::vector< tensor > &input)
	Implements the forward pass. More...

Private Attributes
pooling_params_t	params
const MaxPool2d *	mp_ref

Constructor & Destructor Documentation

MaxUnpool2d()

pytorch::MaxUnpool2d::MaxUnpool2d ( const pooling_params_t &  params, const MaxPool2d *  mp_ref  )

inline

Constructs the MaxUnpool2d layer. Requires pooling parameters that are functionally equivalent to the convolutional parameters.

Parameters

params

Pooling parameters like window, stride, etc.

Member Function Documentation

forward()

std::vector<tensor> pytorch::MaxUnpool2d::forward ( const std::vector< tensor > &  input )

inlinevirtual

Implements the forward pass.

Parameters

input

The input array to be unpooled

Returns

The unpooled array

Implements pytorch::Layer.

operator()()

std::vector<tensor> pytorch::MaxUnpool2d::operator() ( const std::vector< tensor > &  input )

inlinevirtual

Implements the forward pass.

Parameters

input

The input array to be unpooled

Returns

The unpooled array

Implements pytorch::Layer.

Member Data Documentation

mp_ref

const MaxPool2d* pytorch::MaxUnpool2d::mp_ref

private

params

pooling_params_t pytorch::MaxUnpool2d::params

private

---

The documentation for this class was generated from the following file:

• Pooling.hpp