eVision.EasyDeepLearning.EClassificationMetrics Class Reference

Collection of metrics used to evaluate the state of an eVision.EasyDeepLearning.EClassifier. More...

Public Member Functions
None	__init__ (self)
	Constructs an eVision.EasyDeepLearning.EClassificationMetrics object.
None	__init__ (self, eVision.EasyDeepLearning.EClassificationMetrics other)
	Copy constructor of an eVision.EasyDeepLearning.EClassificationMetrics object.
None	Load (self, str path)
	Loads a classification metric.
None	Load (self, eVision.ESerializer|None serializer)
	Loads a classification metric.
None	Save (self, str path)
	Saves a classification metric.
None	Save (self, eVision.ESerializer|None serializer)
	Saves a classification metric.
bool	IsValid (self)
	Indicates whether the object contains at least one classification result.
None	AddResult (self, eVision.EasyDeepLearning.EClassificationResult result, str groundtruthLabel)
	Adds the given result with the corresponding ground truth label to the metrics.
None	AddMetrics (self, eVision.EasyDeepLearning.EClassificationMetrics other)
	Adds the other metrics to the current metrics of this object.
float	GetWeightedAccuracy (self, list[float] weights)
	The label weighted accuracy.
float	GetWeightedAccuracy (self, eVision.EasyDeepLearning.EDataset dataset)
	The label weighted accuracy.
float	GetLabelAccuracy (self, str label)
	The accuracy of the classifier for a given label.
float	GetWeightedError (self, list[float] weights)
	The label weighted error.
float	GetWeightedError (self, eVision.EasyDeepLearning.EDataset dataset)
	The label weighted error.
float	GetLabelError (self, str label)
	The error of the classifier for a given label.
bool	CanComputeWeightedError (self)
	Whether the object can be used to get weighted, balanced, and label errors.
int	GetConfusion (self, str trueClass, str predictedClass)
	Confusion value of one label with another.

Protected Member Functions
float	_Accuracy__getter (self)
	\cond
float	_BalancedAccuracy__getter (self)
	\cond
float	_Error__getter (self)
	\cond
float	_BalancedError__getter (self)
	\cond

Properties
	Accuracy = property(_Accuracy__getter, None)
	\endcond
	BalancedAccuracy = property(_BalancedAccuracy__getter, None)
	\endcond
	Error = property(_Error__getter, None)
	\endcond
	BalancedError = property(_BalancedError__getter, None)
	\endcond

Detailed Description

Collection of metrics used to evaluate the state of an eVision.EasyDeepLearning.EClassifier.

A metric is a value summarizing a collection of classification results (eVision.EasyDeepLearning.EClassificationResult). New results can be added to the object individually with eVision.EasyDeepLearning.EClassificationMetrics.AddResult or collectively with eVision.EasyDeepLearning.EClassificationMetrics.AddMetrics.

eVision.EasyDeepLearning.EClassificationMetrics contains the following metrics:

• the accuracy (see eVision.EasyDeepLearning.EClassificationMetrics.Accuracy)
• the error (see eVision.EasyDeepLearning.EClassificationMetrics.Error)

Remarks:

License(s): BasicTypes

Constructor & Destructor Documentation

__init__() [1/2]

None eVision.EasyDeepLearning.EClassificationMetrics.__init__

(

self

)

Constructs an eVision.EasyDeepLearning.EClassificationMetrics object.

__init__() [2/2]

None eVision.EasyDeepLearning.EClassificationMetrics.__init__	(		self,
		eVision.EasyDeepLearning.EClassificationMetrics	other )

Copy constructor of an eVision.EasyDeepLearning.EClassificationMetrics object.

Parameters

other

Reference to the eVision.EasyDeepLearning.EClassificationMetrics object that should be copied

Member Function Documentation

_Accuracy__getter()

float eVision.EasyDeepLearning.EClassificationMetrics._Accuracy__getter ( self )

protected

\cond

The accuracy of the classifier.

Remarks:

The accuracy is the number of images that were correctly classified (also called the true positives) over the total number of images that was used to evaluate the classifier.

_BalancedAccuracy__getter()

float eVision.EasyDeepLearning.EClassificationMetrics._BalancedAccuracy__getter ( self )

protected

\cond

The balanced accuracy.

Remarks:

The balanced accuracy is the label weighted accuracy with the same weight for each labels.

_BalancedError__getter()

float eVision.EasyDeepLearning.EClassificationMetrics._BalancedError__getter ( self )

protected

\cond

The balanced error.

Remarks:

The balanced error is the label weighted error with the same weight for each labels.

If eVision.EasyDeepLearning.EClassificationMetrics.CanComputeWeightedError is false, this method is an alias for eVision.EasyDeepLearning.EClassificationMetrics.Error.

_Error__getter()

float eVision.EasyDeepLearning.EClassificationMetrics._Error__getter ( self )

protected

\cond

The error of the classifier.

Remarks:

The error, which is also called the loss, is the quantity that is minimized during the training of the deep neural network. For classification, the error is the cross-entropy.

AddMetrics()

None eVision.EasyDeepLearning.EClassificationMetrics.AddMetrics	(		self,
		eVision.EasyDeepLearning.EClassificationMetrics	other )

Adds the other metrics to the current metrics of this object.

Parameters

other

Classification metrics

AddResult()

None eVision.EasyDeepLearning.EClassificationMetrics.AddResult	(		self,
		eVision.EasyDeepLearning.EClassificationResult	result,
		str	groundtruthLabel )

Adds the given result with the corresponding ground truth label to the metrics.

Parameters

result	A reference to an eVision.EasyDeepLearning.EClassificationResult object.
groundtruthLabel	The ground truth label corresponding to the result

CanComputeWeightedError()

bool eVision.EasyDeepLearning.EClassificationMetrics.CanComputeWeightedError

(

self

)

Whether the object can be used to get weighted, balanced, and label errors.

GetConfusion()

int eVision.EasyDeepLearning.EClassificationMetrics.GetConfusion	(		self,
		str	trueClass,
		str	predictedClass )

Confusion value of one label with another.

The confusion value of a label with another is the number of images belonging to this label that are classified as belonging to the other label.

Parameters

trueClass	The label for which to obtain the confusion value
predictedClass	The label with which there is a confusion

GetLabelAccuracy()

float eVision.EasyDeepLearning.EClassificationMetrics.GetLabelAccuracy	(		self,
		str	label )

The accuracy of the classifier for a given label.

Remarks:

The label accuracy is the number of images of a given label that were correctly classified (also called the true positives) over the total number of images of that label that was used to evaluate the classifier.

If there is no results for the given label, the method will throw an exception.

Parameters

label

The label of which we want the accuracy

GetLabelError()

float eVision.EasyDeepLearning.EClassificationMetrics.GetLabelError	(		self,
		str	label )

The error of the classifier for a given label.

Remarks:

The label error corresponds to the error only for images of the given label.

If there is no results for the given label, the method will throw an exception. If eVision.EasyDeepLearning.EClassificationMetrics.CanComputeWeightedError is false, this method is an alias for eVision.EasyDeepLearning.EClassificationMetrics.Error.

Parameters

label

The label from which to get the error.

GetWeightedAccuracy() [1/2]

float eVision.EasyDeepLearning.EClassificationMetrics.GetWeightedAccuracy	(		self,
		eVision.EasyDeepLearning.EDataset	dataset )

The label weighted accuracy.

Remarks:

The weighted accuracy is the weighted average of the label accuracies (see eVision.EasyDeepLearning.EClassificationMetrics.GetLabelAccuracy). If there is no results for a given label, it will be ignored in the weighted accuracy.

Parameters

dataset

Dataset from which to use the label weights

GetWeightedAccuracy() [2/2]

float eVision.EasyDeepLearning.EClassificationMetrics.GetWeightedAccuracy	(		self,
		list[float]	weights )

The label weighted accuracy.

Remarks:

The weighted accuracy is the weighted average of the label accuracies (see eVision.EasyDeepLearning.EClassificationMetrics.GetLabelAccuracy). If there is no results for a given label, it will be ignored in the weighted accuracy.

Parameters

weights

Array of label weights for the labels of the classifier (see eVision.EasyDeepLearning.EDeepLearningTool.GetLabel)

GetWeightedError() [1/2]

float eVision.EasyDeepLearning.EClassificationMetrics.GetWeightedError	(		self,
		eVision.EasyDeepLearning.EDataset	dataset )

The label weighted error.

Remarks:

The weighted error is the weighted average of the label errors (see eVision.EasyDeepLearning.EClassificationMetrics.GetLabelError).

If there isn't any result for a given label, it will be ignored in the weighted error.

If eVision.EasyDeepLearning.EClassificationMetrics.CanComputeWeightedError is false, this method is an alias for eVision.EasyDeepLearning.EClassificationMetrics.Error.

Parameters

dataset

Dataset from which to use the label weights

GetWeightedError() [2/2]

float eVision.EasyDeepLearning.EClassificationMetrics.GetWeightedError	(		self,
		list[float]	weights )

The label weighted error.

Remarks:

The weighted error is the weighted average of the label errors (see eVision.EasyDeepLearning.EClassificationMetrics.GetLabelError).

If there isn't any result for a given label, it will be ignored in the weighted error.

If eVision.EasyDeepLearning.EClassificationMetrics.CanComputeWeightedError is false, this method is an alias for eVision.EasyDeepLearning.EClassificationMetrics.Error.

Parameters

weights

Array of label weights for the labels of the classifier (see eVision.EasyDeepLearning.EDeepLearningTool.GetLabel)

IsValid()

bool eVision.EasyDeepLearning.EClassificationMetrics.IsValid

(

self

)

Indicates whether the object contains at least one classification result.

Load() [1/2]

None eVision.EasyDeepLearning.EClassificationMetrics.Load	(		self,
		eVision.ESerializer | None	serializer )

Loads a classification metric.

The given eVision.ESerializer must have been created for reading.

Parameters

serializer

The serializer.

Load() [2/2]

None eVision.EasyDeepLearning.EClassificationMetrics.Load	(		self,
		str	path )

Loads a classification metric.

The given eVision.ESerializer must have been created for reading.

Parameters

path	The file path.

Save() [1/2]

None eVision.EasyDeepLearning.EClassificationMetrics.Save	(		self,
		eVision.ESerializer | None	serializer )

Saves a classification metric.

The given eVision.ESerializer must have been created for writing.

Parameters

serializer

The eVision.ESerializer object that is written to.

Save() [2/2]

None eVision.EasyDeepLearning.EClassificationMetrics.Save	(		self,
		str	path )

Saves a classification metric.

The given eVision.ESerializer must have been created for writing.

Parameters

path	The file path.

Property Documentation

Accuracy

eVision.EasyDeepLearning.EClassificationMetrics.Accuracy = property(_Accuracy__getter, None)

static

\endcond

The accuracy of the classifier.

Remarks:

The accuracy is the number of images that were correctly classified (also called the true positives) over the total number of images that was used to evaluate the classifier.

BalancedAccuracy

eVision.EasyDeepLearning.EClassificationMetrics.BalancedAccuracy = property(_BalancedAccuracy__getter, None)

static

\endcond

The balanced accuracy.

Remarks:

The balanced accuracy is the label weighted accuracy with the same weight for each labels.

BalancedError

eVision.EasyDeepLearning.EClassificationMetrics.BalancedError = property(_BalancedError__getter, None)

static

\endcond

The balanced error.

Remarks:

The balanced error is the label weighted error with the same weight for each labels.

If eVision.EasyDeepLearning.EClassificationMetrics.CanComputeWeightedError is false, this method is an alias for eVision.EasyDeepLearning.EClassificationMetrics.Error.

Error

eVision.EasyDeepLearning.EClassificationMetrics.Error = property(_Error__getter, None)

static

\endcond

The error of the classifier.

Remarks:

The error, which is also called the loss, is the quantity that is minimized during the training of the deep neural network. For classification, the error is the cross-entropy.