eVision.EasyDeepLearning.EDeepOCRToolMetrics Class Reference

Metrics for EDeepOCRTool. More...

Inheritance diagram for eVision.EasyDeepLearning.EDeepOCRToolMetrics:

Collaboration diagram for eVision.EasyDeepLearning.EDeepOCRToolMetrics:

Public Member Functions
	EDeepOCRToolMetrics ()
	Constructs a EDeepOCRToolMetrics.
	EDeepOCRToolMetrics (eVision.EasyDeepLearning.EDeepOCRToolMetrics other)
	Copy constructor of a EDeepOCRToolMetrics object.
void	AddResult (eVision.EasyDeepLearning.EDeepOCRToolResult result)
	Add a result to the metrics.
void	RemoveResult (eVision.EasyDeepLearning.EDeepOCRToolResult result)
	Removes a result from the metrics.
void	AddResult (float loss)
	Adds a loss result to the metrics.
void	RemoveResult (float loss)
	Removes a loss result from the metrics.
bool	IsValid ()
	Whether the metrics contains data or not.
float	ComputeLocationThresholdMaximizingRecall ()
	Computes the location threshold that will maximize the EDeepOCRToolMetrics.LocationRecall.
float	ComputeLocationThresholdMaximizingFScore ()
	Computes the location threshold that will maximize the EDeepOCRToolMetrics.LocationFScore.
float	ComputeLocationThresholdMaximizingPrecision ()
	Computes the location threshold that will maximize the EDeepOCRToolMetrics.LocationPrecision.
void	Load (string path)
	Loads the EDeepOCRToolMetrics. The given ESerializer must have been created for reading.
void	Load (eVision.ESerializer serializer)
	Loads the EDeepOCRToolMetrics. The given ESerializer must have been created for reading.
void	Save (string path)
	Saves the EDeepOCRToolMetrics. The given ESerializer must have been created for writing.
void	Save (eVision.ESerializer serializer)
	Saves the EDeepOCRToolMetrics. The given ESerializer must have been created for writing.

Properties
eVision.EasyDeepLearning.EDeepOCRModel	Model [get]
	The type of model for this result.
float	Error [get]
	Error computed during training.
float	LocationFScore [get]
	F Score of the location (harmonic mean of the precision and recall).
float	LocationRecall [get]
	Recall of the location (proportion of ground truth text that are correctly detected).
float	LocationPrecision [get]
	Precision of the location (proportion of detected texts that are correctly detected).
eVision.EasyDeepLearning.EClassificationMetrics	OrientationMetrics [get]
	Get the metrics for the orientation.
float	RecognitionImageAccuracy [get]
	Get the image-wise accuracy (mean per-image accuracy).
float	RecognitionWordAccuracy [get]
	Get the word-wise accuracy (share of words that were correctly read in all images). Biased towards images containing a large number of words.
float	LocationThreshold [get, set]
	Location threshold for which to compute the metrics.

Detailed Description

Metrics for EDeepOCRTool.

License(s): EasyDeepOCR

Constructor & Destructor Documentation

EDeepOCRToolMetrics() [1/2]

eVision.EasyDeepLearning.EDeepOCRToolMetrics.EDeepOCRToolMetrics

(

)

Constructs a EDeepOCRToolMetrics.

EDeepOCRToolMetrics() [2/2]

eVision.EasyDeepLearning.EDeepOCRToolMetrics.EDeepOCRToolMetrics

(

eVision.EasyDeepLearning.EDeepOCRToolMetrics

other

)

Copy constructor of a EDeepOCRToolMetrics object.

Parameters

other

Reference to the EDeepOCRToolMetrics object that should be copied

Member Function Documentation

AddResult() [1/2]

void eVision.EasyDeepLearning.EDeepOCRToolMetrics.AddResult

(

eVision.EasyDeepLearning.EDeepOCRToolResult

result

)

Add a result to the metrics.

Parameters

result

Result to add to the metrics.

AddResult() [2/2]

void eVision.EasyDeepLearning.EDeepOCRToolMetrics.AddResult

(

float

loss

)

Adds a loss result to the metrics.

Parameters

loss	Loss to add to the metrics.

ComputeLocationThresholdMaximizingFScore()

float eVision.EasyDeepLearning.EDeepOCRToolMetrics.ComputeLocationThresholdMaximizingFScore

(

)

Computes the location threshold that will maximize the EDeepOCRToolMetrics.LocationFScore.

ComputeLocationThresholdMaximizingPrecision()

float eVision.EasyDeepLearning.EDeepOCRToolMetrics.ComputeLocationThresholdMaximizingPrecision

(

)

Computes the location threshold that will maximize the EDeepOCRToolMetrics.LocationPrecision.

ComputeLocationThresholdMaximizingRecall()

float eVision.EasyDeepLearning.EDeepOCRToolMetrics.ComputeLocationThresholdMaximizingRecall

(

)

Computes the location threshold that will maximize the EDeepOCRToolMetrics.LocationRecall.

IsValid()

bool eVision.EasyDeepLearning.EDeepOCRToolMetrics.IsValid

(

)

Whether the metrics contains data or not.

Load() [1/2]

void eVision.EasyDeepLearning.EDeepOCRToolMetrics.Load

(

eVision.ESerializer

serializer

)

Loads the EDeepOCRToolMetrics. The given ESerializer must have been created for reading.

Parameters

serializer

The serializer.

Load() [2/2]

void eVision.EasyDeepLearning.EDeepOCRToolMetrics.Load

(

string

path

)

Loads the EDeepOCRToolMetrics. The given ESerializer must have been created for reading.

Parameters

path	The file path.

RemoveResult() [1/2]

void eVision.EasyDeepLearning.EDeepOCRToolMetrics.RemoveResult

(

eVision.EasyDeepLearning.EDeepOCRToolResult

result

)

Removes a result from the metrics.

Parameters

result

Result to remove from the metrics.

RemoveResult() [2/2]

void eVision.EasyDeepLearning.EDeepOCRToolMetrics.RemoveResult

(

float

loss

)

Removes a loss result from the metrics.

Parameters

loss	Loss result to remove from the metrics.

Save() [1/2]

void eVision.EasyDeepLearning.EDeepOCRToolMetrics.Save

(

eVision.ESerializer

serializer

)

Saves the EDeepOCRToolMetrics. The given ESerializer must have been created for writing.

Parameters

serializer

The serializer.

Save() [2/2]

void eVision.EasyDeepLearning.EDeepOCRToolMetrics.Save

(

string

path

)

Saves the EDeepOCRToolMetrics. The given ESerializer must have been created for writing.

Parameters

path	The file path.

Property Documentation

Error

float eVision.EasyDeepLearning.EDeepOCRToolMetrics.Error

get

Error computed during training.

LocationFScore

float eVision.EasyDeepLearning.EDeepOCRToolMetrics.LocationFScore

get

F Score of the location (harmonic mean of the precision and recall).

LocationPrecision

float eVision.EasyDeepLearning.EDeepOCRToolMetrics.LocationPrecision

get

Precision of the location (proportion of detected texts that are correctly detected).

LocationRecall

float eVision.EasyDeepLearning.EDeepOCRToolMetrics.LocationRecall

get

Recall of the location (proportion of ground truth text that are correctly detected).

LocationThreshold

float eVision.EasyDeepLearning.EDeepOCRToolMetrics.LocationThreshold

getset

Location threshold for which to compute the metrics.

Model

eVision.EasyDeepLearning.EDeepOCRModel eVision.EasyDeepLearning.EDeepOCRToolMetrics.Model

get

The type of model for this result.

OrientationMetrics

eVision.EasyDeepLearning.EClassificationMetrics eVision.EasyDeepLearning.EDeepOCRToolMetrics.OrientationMetrics

get

Get the metrics for the orientation.

RecognitionImageAccuracy

float eVision.EasyDeepLearning.EDeepOCRToolMetrics.RecognitionImageAccuracy

get

Get the image-wise accuracy (mean per-image accuracy).

RecognitionWordAccuracy

float eVision.EasyDeepLearning.EDeepOCRToolMetrics.RecognitionWordAccuracy

get

Get the word-wise accuracy (share of words that were correctly read in all images). Biased towards images containing a large number of words.