TestBackpropagation.h File Reference

#include <iostream>
#include "TMVA/DNN/Functions.h"
#include "TMVA/DNN/Net.h"
#include "Utility.h"

Include dependency graph for TestBackpropagation.h:

This graph shows which files directly or indirectly include this file:

Functions
template<typename Architecture >
auto	evaluate_net_bias (TNet< Architecture > &net, typename Architecture::Matrix_t &X, const typename Architecture::Matrix_t &Y, size_t l, size_t i, typename Architecture::Scalar_t dx) -> typename Architecture::Scalar_t
	Compute the loss of the net as a function of the weight at index i in layer l. More...
template<typename Architecture >
auto	evaluate_net_weight (TNet< Architecture > &net, typename Architecture::Matrix_t &X, const typename Architecture::Matrix_t &Y, size_t l, size_t i, size_t j, typename Architecture::Scalar_t dx) -> typename Architecture::Scalar_t
	Compute the loss of the net as a function of the weight at index (i,j) in layer l. More...
template<typename Architecture >
auto	testBackpropagationBiasesLinear (typename Architecture::Scalar_t dx) -> typename Architecture::Scalar_t
	Generate a random net, perform forward and backward propagation and check the bias gradients using numerical differentiation. More...
template<typename Architecture >
auto	testBackpropagationL1Regularization (typename Architecture::Scalar_t dx) -> typename Architecture::Scalar_t
	Generate a random, linear net, perform forward and backward propagation with L1 regularization and check the weight gradients using numerical differentiation. More...
template<typename Architecture >
auto	testBackpropagationL2Regularization (typename Architecture::Scalar_t dx) -> typename Architecture::Scalar_t
	Generate a random, linear net, perform forward and backward propagation with L2 regularization and check the weight gradients using numerical differentiation. More...
template<typename Architecture >
auto	testBackpropagationWeightsLinear (typename Architecture::Scalar_t dx) -> typename Architecture::Scalar_t
	Generate a random net, perform forward and backward propagation and check the weight gradients using numerical differentiation. More...

Function Documentation

evaluate_net_bias()

template<typename Architecture >

auto evaluate_net_bias	(	TNet< Architecture > &	net,
		typename Architecture::Matrix_t &	X,
		const typename Architecture::Matrix_t &	Y,
		size_t	l,
		size_t	i,
		typename Architecture::Scalar_t	dx
	)		-> typename Architecture::Scalar_t

Compute the loss of the net as a function of the weight at index i in layer l.

dx is added as an offset to the current value of the weight.

Definition at line 56 of file TestBackpropagation.h.

evaluate_net_weight()

template<typename Architecture >

auto evaluate_net_weight	(	TNet< Architecture > &	net,
		typename Architecture::Matrix_t &	X,
		const typename Architecture::Matrix_t &	Y,
		size_t	l,
		size_t	i,
		size_t	j,
		typename Architecture::Scalar_t	dx
	)		-> typename Architecture::Scalar_t

Compute the loss of the net as a function of the weight at index (i,j) in layer l.

dx is added as an offset to the current value of the weight.

Definition at line 35 of file TestBackpropagation.h.

testBackpropagationBiasesLinear()

template<typename Architecture >

auto testBackpropagationBiasesLinear

(

typename Architecture::Scalar_t

dx

)

-> typename Architecture::Scalar_t

Generate a random net, perform forward and backward propagation and check the bias gradients using numerical differentiation.

Returns the maximum relative gradient error and also prints it to stdout.

Definition at line 298 of file TestBackpropagation.h.

testBackpropagationL1Regularization()

template<typename Architecture >

auto testBackpropagationL1Regularization

(

typename Architecture::Scalar_t

dx

)

-> typename Architecture::Scalar_t

Generate a random, linear net, perform forward and backward propagation with L1 regularization and check the weight gradients using numerical differentiation.

Returns the maximum relative gradient error and also prints it to stdout.

Definition at line 151 of file TestBackpropagation.h.

testBackpropagationL2Regularization()

template<typename Architecture >

auto testBackpropagationL2Regularization

(

typename Architecture::Scalar_t

dx

)

-> typename Architecture::Scalar_t

Generate a random, linear net, perform forward and backward propagation with L2 regularization and check the weight gradients using numerical differentiation.

Returns the maximum relative gradient error and also prints it to stdout.

Definition at line 226 of file TestBackpropagation.h.

testBackpropagationWeightsLinear()

template<typename Architecture >

auto testBackpropagationWeightsLinear

(

typename Architecture::Scalar_t

dx

)

-> typename Architecture::Scalar_t

Generate a random net, perform forward and backward propagation and check the weight gradients using numerical differentiation.

Returns the maximum relative gradient error and also prints it to stdout.

Definition at line 77 of file TestBackpropagation.h.