ROOT 6.14/05 Reference Guide |
LayerData holds the data of one layer.
LayerData holds the data of one layer, but not its layout
Definition at line 437 of file NeuralNet.h.
Public Types | |
typedef DropContainer::const_iterator | const_dropout_iterator |
typedef function_container_type::const_iterator | const_function_iterator_type |
typedef container_type::const_iterator | const_iterator_type |
typedef std::vector< double > | container_type |
typedef std::vector< std::function< double(double)> > | function_container_type |
typedef function_container_type::iterator | function_iterator_type |
typedef container_type::iterator | iterator_type |
Public Member Functions | |
LayerData (const_iterator_type itInputBegin, const_iterator_type itInputEnd, ModeOutputValues eModeOutput=ModeOutputValues::DIRECT) | |
c'tor of LayerData More... | |
LayerData (size_t inputSize) | |
c'tor of LayerData More... | |
LayerData (size_t size, const_iterator_type itWeightBegin, iterator_type itGradientBegin, std::shared_ptr< std::function< double(double)>> activationFunction, std::shared_ptr< std::function< double(double)>> inverseActivationFunction, ModeOutputValues eModeOutput=ModeOutputValues::DIRECT) | |
c'tor of LayerData More... | |
LayerData (size_t size, const_iterator_type itWeightBegin, std::shared_ptr< std::function< double(double)>> activationFunction, ModeOutputValues eModeOutput=ModeOutputValues::DIRECT) | |
c'tor of LayerData More... | |
LayerData (const LayerData &other) | |
copy c'tor of LayerData More... | |
LayerData (LayerData &&other) | |
move c'tor of LayerData More... | |
~LayerData () | |
std::shared_ptr< std::function< double(double)> > | activationFunction () const |
void | clear () |
clear the values and the deltas More... | |
void | clearDropOut () |
clear the drop-out-data for this layer More... | |
iterator_type | deltasBegin () |
returns iterator to the begin of the deltas (back-propagation) More... | |
const_iterator_type | deltasBegin () const |
returns const iterator to the begin of the deltas (back-propagation) More... | |
iterator_type | deltasEnd () |
returns iterator to the end of the deltas (back-propagation) More... | |
const_iterator_type | deltasEnd () const |
returns const iterator to the end of the deltas (back-propagation) More... | |
const_dropout_iterator | dropOut () const |
return the begin of the drop-out information More... | |
iterator_type | gradientsBegin () |
returns iterator to the begin of the gradients More... | |
const_iterator_type | gradientsBegin () const |
returns const iterator to the begin of the gradients More... | |
bool | hasDropOut () const |
has this layer drop-out turned on? More... | |
std::shared_ptr< std::function< double(double)> > | inverseActivationFunction () const |
ModeOutputValues | outputMode () const |
returns the output mode More... | |
container_type | probabilities () const |
computes the probabilities from the current node values and returns them More... | |
template<typename Iterator > | |
void | setDropOut (Iterator itDrop) |
set the drop-out info for this layer More... | |
void | setInput (const_iterator_type itInputBegin, const_iterator_type itInputEnd) |
change the input iterators More... | |
size_t | size () const |
return the size of the layer More... | |
iterator_type | valueGradientsBegin () |
returns iterator to the begin of the gradients of the node values More... | |
const_iterator_type | valueGradientsBegin () const |
returns const iterator to the begin of the gradients More... | |
iterator_type | valueGradientsEnd () |
returns iterator to the end of the gradients of the node values More... | |
const_iterator_type | valueGradientsEnd () const |
returns const iterator to the end of the gradients More... | |
const_iterator_type | valuesBegin () const |
returns const iterator to the begin of the (node) values More... | |
iterator_type | valuesBegin () |
returns iterator to the begin of the (node) values More... | |
const_iterator_type | valuesEnd () const |
returns iterator to the end of the (node) values More... | |
iterator_type | valuesEnd () |
returns iterator to the end of the (node) values More... | |
const_iterator_type | weightsBegin () const |
returns const iterator to the begin of the weights for this layer More... | |
Private Member Functions | |
container_type | computeProbabilities () const |
compute the probabilities from the node values More... | |
Private Attributes | |
std::shared_ptr< std::function< double(double)> > | m_activationFunction |
activation function for this layer More... | |
std::vector< double > | m_deltas |
stores the deltas for the DNN training More... | |
ModeOutputValues | m_eModeOutput |
stores the output mode (DIRECT, SIGMOID, SOFTMAX) More... | |
bool | m_hasDropOut |
dropOut is turned on? More... | |
bool | m_hasGradients |
does this layer have gradients (only if in training mode) More... | |
bool | m_hasWeights |
does this layer have weights (it does not if it is the input layer) More... | |
std::shared_ptr< std::function< double(double)> > | m_inverseActivationFunction |
inverse activation function for this layer More... | |
bool | m_isInputLayer |
is this layer an input layer More... | |
const_iterator_type | m_itConstWeightBegin |
const iterator to the first weight of this layer in the weight vector More... | |
const_dropout_iterator | m_itDropOut |
iterator to a container indicating if the corresponding node is to be dropped More... | |
iterator_type | m_itGradientBegin |
iterator to the first gradient of this layer in the gradient vector More... | |
const_iterator_type | m_itInputBegin |
iterator to the first of the nodes in the input node vector More... | |
const_iterator_type | m_itInputEnd |
iterator to the end of the nodes in the input node vector More... | |
size_t | m_size |
std::vector< double > | m_valueGradients |
stores the gradients of the values (nodes) More... | |
std::vector< double > | m_values |
stores the values of the nodes in this layer More... | |
#include <TMVA/NeuralNet.h>
typedef DropContainer::const_iterator TMVA::DNN::LayerData::const_dropout_iterator |
Definition at line 449 of file NeuralNet.h.
typedef function_container_type::const_iterator TMVA::DNN::LayerData::const_function_iterator_type |
Definition at line 447 of file NeuralNet.h.
typedef container_type::const_iterator TMVA::DNN::LayerData::const_iterator_type |
Definition at line 443 of file NeuralNet.h.
typedef std::vector<double> TMVA::DNN::LayerData::container_type |
Definition at line 440 of file NeuralNet.h.
typedef std::vector<std::function<double(double)> > TMVA::DNN::LayerData::function_container_type |
Definition at line 445 of file NeuralNet.h.
typedef function_container_type::iterator TMVA::DNN::LayerData::function_iterator_type |
Definition at line 446 of file NeuralNet.h.
typedef container_type::iterator TMVA::DNN::LayerData::iterator_type |
Definition at line 442 of file NeuralNet.h.
TMVA::DNN::LayerData::LayerData | ( | const_iterator_type | itInputBegin, |
const_iterator_type | itInputEnd, | ||
ModeOutputValues | eModeOutput = ModeOutputValues::DIRECT |
||
) |
c'tor of LayerData
C'tor of LayerData for the input layer
itInputBegin | iterator to the begin of a vector which holds the values of the nodes of the neural net |
itInputEnd | iterator to the end of a vector which holdsd the values of the nodes of the neural net |
eModeOutput | indicates a potential tranformation of the output values before further computation DIRECT does not further transformation; SIGMOID applies a sigmoid transformation to each output value (to create a probability); SOFTMAX applies a softmax transformation to all output values (mutually exclusive probability) |
Definition at line 81 of file NeuralNet.cxx.
TMVA::DNN::LayerData::LayerData | ( | size_t | inputSize | ) |
c'tor of LayerData
C'tor of LayerData for the input layer
inputSize | input size of this layer |
Definition at line 68 of file NeuralNet.cxx.
|
inline |
Definition at line 472 of file NeuralNet.h.
TMVA::DNN::LayerData::LayerData | ( | size_t | size, |
const_iterator_type | itWeightBegin, | ||
iterator_type | itGradientBegin, | ||
std::shared_ptr< std::function< double(double)>> | activationFunction, | ||
std::shared_ptr< std::function< double(double)>> | inverseActivationFunction, | ||
ModeOutputValues | eModeOutput = ModeOutputValues::DIRECT |
||
) |
c'tor of LayerData
C'tor of LayerData for all layers which are not the input layer; Used during the training of the DNN
size | size of the layer |
itWeightBegin | indicates the start of the weights for this layer on the weight vector |
itGradientBegin | indicates the start of the gradients for this layer on the gradient vector |
itFunctionBegin | indicates the start of the vector of activation functions for this layer on the activation function vector |
itInverseFunctionBegin | indicates the start of the vector of activation functions for this layer on the activation function vector |
eModeOutput | indicates a potential tranformation of the output values before further computation DIRECT does not further transformation; SIGMOID applies a sigmoid transformation to each output value (to create a probability); SOFTMAX applies a softmax transformation to all output values (mutually exclusive probability) |
Definition at line 97 of file NeuralNet.cxx.
TMVA::DNN::LayerData::LayerData | ( | size_t | size, |
const_iterator_type | itWeightBegin, | ||
std::shared_ptr< std::function< double(double)>> | activationFunction, | ||
ModeOutputValues | eModeOutput = ModeOutputValues::DIRECT |
||
) |
c'tor of LayerData
C'tor of LayerData for all layers which are not the input layer; Used during the application of the DNN
size | size of the layer |
itWeightBegin | indicates the start of the weights for this layer on the weight vector |
itFunctionBegin | indicates the start of the vector of activation functions for this layer on the activation function vector |
eModeOutput | indicates a potential tranformation of the output values before further computation DIRECT does not further transformation; SIGMOID applies a sigmoid transformation to each output value (to create a probability); SOFTMAX applies a softmax transformation to all output values (mutually exclusive probability) |
Definition at line 122 of file NeuralNet.cxx.
|
inline |
copy c'tor of LayerData
Definition at line 519 of file NeuralNet.h.
|
inline |
move c'tor of LayerData
Definition at line 542 of file NeuralNet.h.
|
inline |
Definition at line 611 of file NeuralNet.h.
|
inline |
clear the values and the deltas
Definition at line 580 of file NeuralNet.h.
|
inline |
clear the drop-out-data for this layer
Definition at line 624 of file NeuralNet.h.
|
private |
compute the probabilities from the node values
Definition at line 140 of file NeuralNet.cxx.
|
inline |
returns iterator to the begin of the deltas (back-propagation)
Definition at line 595 of file NeuralNet.h.
|
inline |
returns const iterator to the begin of the deltas (back-propagation)
Definition at line 598 of file NeuralNet.h.
|
inline |
returns iterator to the end of the deltas (back-propagation)
Definition at line 596 of file NeuralNet.h.
|
inline |
returns const iterator to the end of the deltas (back-propagation)
Definition at line 599 of file NeuralNet.h.
|
inline |
return the begin of the drop-out information
Definition at line 627 of file NeuralNet.h.
|
inline |
returns iterator to the begin of the gradients
Definition at line 607 of file NeuralNet.h.
|
inline |
returns const iterator to the begin of the gradients
Definition at line 608 of file NeuralNet.h.
|
inline |
has this layer drop-out turned on?
Definition at line 626 of file NeuralNet.h.
|
inline |
Definition at line 612 of file NeuralNet.h.
|
inline |
returns the output mode
Definition at line 592 of file NeuralNet.h.
|
inline |
computes the probabilities from the current node values and returns them
Definition at line 593 of file NeuralNet.h.
|
inline |
set the drop-out info for this layer
Definition at line 618 of file NeuralNet.h.
|
inline |
change the input iterators
itInputBegin | indicates the start of the input node vector |
itInputEnd | indicates the end of the input node vector |
Definition at line 569 of file NeuralNet.h.
|
inline |
return the size of the layer
Definition at line 629 of file NeuralNet.h.
|
inline |
returns iterator to the begin of the gradients of the node values
Definition at line 601 of file NeuralNet.h.
|
inline |
returns const iterator to the begin of the gradients
Definition at line 604 of file NeuralNet.h.
|
inline |
returns iterator to the end of the gradients of the node values
Definition at line 602 of file NeuralNet.h.
|
inline |
returns const iterator to the end of the gradients
Definition at line 605 of file NeuralNet.h.
|
inline |
returns const iterator to the begin of the (node) values
Definition at line 586 of file NeuralNet.h.
|
inline |
returns iterator to the begin of the (node) values
Definition at line 589 of file NeuralNet.h.
|
inline |
returns iterator to the end of the (node) values
Definition at line 587 of file NeuralNet.h.
|
inline |
returns iterator to the end of the (node) values
Definition at line 590 of file NeuralNet.h.
|
inline |
returns const iterator to the begin of the weights for this layer
Definition at line 609 of file NeuralNet.h.
|
private |
activation function for this layer
Definition at line 655 of file NeuralNet.h.
|
private |
stores the deltas for the DNN training
Definition at line 646 of file NeuralNet.h.
|
private |
stores the output mode (DIRECT, SIGMOID, SOFTMAX)
Definition at line 662 of file NeuralNet.h.
|
private |
dropOut is turned on?
Definition at line 650 of file NeuralNet.h.
|
private |
does this layer have gradients (only if in training mode)
Definition at line 660 of file NeuralNet.h.
|
private |
does this layer have weights (it does not if it is the input layer)
Definition at line 659 of file NeuralNet.h.
|
private |
inverse activation function for this layer
Definition at line 656 of file NeuralNet.h.
|
private |
is this layer an input layer
Definition at line 658 of file NeuralNet.h.
|
private |
const iterator to the first weight of this layer in the weight vector
Definition at line 652 of file NeuralNet.h.
|
private |
iterator to a container indicating if the corresponding node is to be dropped
Definition at line 649 of file NeuralNet.h.
|
private |
iterator to the first gradient of this layer in the gradient vector
Definition at line 653 of file NeuralNet.h.
|
private |
iterator to the first of the nodes in the input node vector
Definition at line 643 of file NeuralNet.h.
|
private |
iterator to the end of the nodes in the input node vector
Definition at line 644 of file NeuralNet.h.
|
private |
Definition at line 641 of file NeuralNet.h.
|
private |
stores the gradients of the values (nodes)
Definition at line 647 of file NeuralNet.h.
|
private |
stores the values of the nodes in this layer
Definition at line 648 of file NeuralNet.h.