eVision.EasyDeepLearning Namespace Reference

Classes
class	EClassificationDataset
	EClassificationDataset manages a dataset of images. More...
class	EClassificationMetrics
	Collection of metrics used to evaluate the state of an EClassifier. More...
class	EClassificationResult
	An EClassificationResult object represents the result of a classification. More...
class	EClassifier
	EClassifier allows to train a classifier using an EDataset object and classify new images. More...
class	EDataAugmentation
	An EDataAugmentation object is responsible for storing the data augmentation parameters and generating new transformed images from existing ones. Deep learning algorithms are not invariant to the location, scale, or rotation of the elements of interest in the image. Thus, if the final application requires the deep learning algorithm to be invariant to those characteristics, the dataset must contain images covering the spectrum of those characteristics. More...
class	EDataset
	EDataset manages a dataset of images. More...
class	EDatasetSplit
	The EDatasetSplit maps the image indexes of a EDataset to a dataset split type (EDatasetType). More...
class	EDeepLearningBenchmark
	Deep Learning benchmark results. More...
class	EDeepLearningBenchmarkSettings
	Class representing the settings of a Deep Learning benchmark. More...
class	EDeepLearningDefectDetectionMetrics
	Collection of metrics used to evaluate a defect detection problem where the detection is based on the thresholding of a score produced by the underlying deep learning tool, e.g. a EUnsupervisedSegmenter tool or a ESupervisedSegmenter tool. More...
class	EDeepLearningDevice
	Class representing a device that can run a neural network. A device is fully described by its EDeepLearningDevice.Name and EDeepLearningDevice.EngineName. More...
class	EDeepLearningExecutionSettings
	Execution settings for a Deep Learning tool. More...
class	EDeepLearningProject
	Deep learning project. More...
class	EDeepLearningTool
	EDeepLearningTool represents the common operations of deep learning tools. More...
class	EDeepOCRModelParameters
	Base class for the parameters of EasyDeepOCR models for user retraining. More...
class	EDeepOCRModelParametersTextLocation
	Parameters for the user retraining of the text location model of EasyDeepOCR. More...
class	EDeepOCRTool
	Trains the EasyDeepOCR Deep Learning models. More...
class	EDeepOCRToolMetrics
	Metrics for EDeepOCRTool. More...
class	EDeepOCRToolResult
	Result for EDeepOCRTool. More...
class	EInterestPointLocator
	EasyLocate tool - interest point version. More...
class	ELocator
	EasyLocate tool - axis aligned bounding box version. More...
class	ELocatorBase
	Base class for EasyLocate tool. More...
class	ELocatorMetrics
	Collection of metrics used to evaluate the results of a ELocator tool. More...
class	ELocatorMetricsGenerator
	Locator metric generator. More...
class	ELocatorObject
	Object for a ELocator tool. More...
class	ELocatorPredictedObject
	Object predicted by a ELocator tool. More...
class	ELocatorResult
	Result of a ELocator tool. More...
struct	EROCPoint
	The structure representing a point on the ROC (Receiver Operating Characteristic) curve. More...
class	ESupervisedSegmenter
	Supervised segmentation tool. More...
class	ESupervisedSegmenterBlob
	A blob detected by a supervised segmentation tool (see ESupervisedSegmenter and ESupervisedSegmenterResult). More...
class	ESupervisedSegmenterMetrics
	Collection of metrics used to evaluate the state of a ESupervisedSegmenter tool. More...
class	ESupervisedSegmenterResult
	An ESupervisedSegmenterResult object represents the result of a supervised segmenter. More...
class	EUnsupervisedSegmenter
	Unsupervised segmentation tool. More...
class	EUnsupervisedSegmenterMetrics
	Collection of metrics used to evaluate the state of an EUnsupervisedSegmenter. More...
class	EUnsupervisedSegmenterResult
	An EUnsupervisedSegmenterResult object represents the result of a EUnsupervisedSegmenter tool. More...
struct	EWeightedPrecisionRecallPoint
	Weighted precision and recall for a given threshold. More...

Enumerations
enum	EClassifierCapacity : int { Small = 0 , Normal = 1 , Large = 2 , Onnx = 3 }
	The capacity of the classifier network. More...
enum	EConfusionMatrixElement : int { CorrectlyClassifiedGoodSample = 0 , BadlyClassifiedGoodSample = 1 , CorrectlyClassifiedDefectiveSample = 2 , BadlyClassifiedDefectiveSample = 3 }
	The various elements representing the cases of the 2x2 confusion matrix used in EDeepLearningDefectDetectionMetrics. More...
enum	EDatasetType : int { Training = 1 , Validation = 2 , Test = 4 , All = 7 }
	Type of dataset split. More...
enum	EDeepLearningDeviceType : int { CPU = 0 , GPU = 1 , Other = 2 , NotADevice = 3 }
	Different types of Deep Learning device. More...
enum	EDeepLearningInferencePrecision : int { INT8 = 0 , FLOAT16 = 1 , FLOAT32 = 2 }
	Precisions supported by a Deep Learning device. More...
enum	EDeepLearningLearningRateType : int {   Standard = 0 , Fixed = 1 , Custom = 2 , Decay = 3 ,   WarmRestart = 4 }
	The type of learning rate indicates how the learning rate changes during training. More...
enum	EDeepLearningToolType : int {   None = 0 , EasyClassify = 1 , EasySegmentUnsupervised = 2 , EasySegmentSupervised = 3 ,   EasyLocate = 4 , EasyLocateInterestPoint = 5 , DeepOCR = 6 }
	Type of EasyDeepLearning tool. More...
enum	EDeepOCRModel : int { Location = 0 , Orientation = 1 , Recognition = 2 , None = 3 }
	EasyDeepOCR models that are trainable with EDeepOCRTool. More...
enum	EDLDataAugmentationType : int { Random = 0 , LowerLimit = 1 , UpperLimit = 2 }
	This enum represents how the data augmentation transformation are generated. More...
enum	EHeatmapColormap : int { YellowToRed = 0 , Jet = 1 , Grayscale = 2 }
	Color maps for visualizing the heatmap. More...
enum	EImageAnnotationFormat : int { None = 0 , YOLOTXT = 1 , PASCALVOCXML = 2 }
	Supported formats of annotation files for image file. More...
enum	ELocatorCapacity : int { Small = 0 , Normal = 1 , Large = 2 }
	The capacity of the locator deep learning network. More...
enum	ELocatorFeature : int { None = 0 , Position = 1 , Size = 2 }
	Features supported by the ELocator or ELocatorObject. More...
enum	ESupervisedSegmenterCapacity : int { Small = 0 , Normal = 1 , Large = 2 }
	The capacity of the supervised segmentation network. More...
enum	EUnsupervisedSegmenterCapacity : int { Small = 0 , Normal = 1 , Large = 2 }
	The capacity of the unsupervised network. More...

Enumeration Type Documentation

EClassifierCapacity

enum eVision.EasyDeepLearning.EClassifierCapacity : int

The capacity of the classifier network.

A larger capacity means that the underlying neural network is capable of learning more information but it will be slower. Also a lower capacity allow for a smaller minimal height and width of input.

Deprecated

Deprecated in version 26.10. Scheduled to be removed in version 27.6.

Enumerator
Small	Small capacity.
Normal	Normal capacity.
Large	Large capacity.
Onnx	Onnx classifier capacity.

EConfusionMatrixElement

enum eVision.EasyDeepLearning.EConfusionMatrixElement : int

The various elements representing the cases of the 2x2 confusion matrix used in EDeepLearningDefectDetectionMetrics.

Enumerator
CorrectlyClassifiedGoodSample	A CorrectlyClassifiedGoodSample is a good images classified as good. It is also called a true negative.
BadlyClassifiedGoodSample	A BadlyClassifiedGoodSample is a good images classified as defective. It is also called a false positive.
CorrectlyClassifiedDefectiveSample	A CorrectlyClassifiedDefectiveSample is a defective images classified as defective. It is also called a true positive.
BadlyClassifiedDefectiveSample	A BadlyClassifiedDefectiveSample is a good images classified as good. It is also called a false negative.

EDatasetType

enum eVision.EasyDeepLearning.EDatasetType : int

Type of dataset split.

Enumerator
Training	Training dataset (images used to train the model).
Validation	Validation dataset (images used to select the model ith the best performance during training).
Test	Test dataset (images not used during training).
All	Training, validation, or test dataset.

EDeepLearningDeviceType

enum eVision.EasyDeepLearning.EDeepLearningDeviceType : int

Different types of Deep Learning device.

Enumerator
CPU	The Deep Learning device is a CPU.
GPU	The Deep Learning device is a GPU.
Other	The Deep Learning device is neither a CPU nor a GPU.
NotADevice	The Deep Learning device is not a device.

EDeepLearningInferencePrecision

enum eVision.EasyDeepLearning.EDeepLearningInferencePrecision : int

Precisions supported by a Deep Learning device.

Enumerator
INT8	8 bit integer precision.
FLOAT16	16 bit floating point precision.
FLOAT32	32 bit floating point precision.

EDeepLearningLearningRateType

enum eVision.EasyDeepLearning.EDeepLearningLearningRateType : int

The type of learning rate indicates how the learning rate changes during training.

Enumerator
Standard	The learning rate is reduced after each training iterations where the training error has not significantly changed. Parameters: None
Fixed	The learning rate value remains fixed with the value given as parameter during the training. Parameters= [lr]
Custom	The learning rate is set for each iterations at the corresponding value in the vector of parameters. Parameters: [lr0, lr1, ..., lrN]. The last learning rate in the parameters will be used when the number of training iterations is bigger than the vector of parameters.
Decay	The learning rate is reduced by a factor after each iterations until a minimum. Parameters: [Start learning rate, Minimum learning rate, Decay factor]
WarmRestart	The learning rate goes from maxLr to minLr during the n-th period of periodN iterations. The length of the first period is given by the parameter period0 and is multiplied by the factor f after each period. Thus, periodN = f^n * period0. Parameters: [minMr, maxLr, period0, f]

EDeepLearningToolType

enum eVision.EasyDeepLearning.EDeepLearningToolType : int

Type of EasyDeepLearning tool.

Enumerator
None	Not an EasyDeepLearning tool.
EasyClassify	EasyClassify tool.
EasySegmentUnsupervised	EasySegmentUnsupervised tool.
EasySegmentSupervised	EasySegmentSupervised tool.
EasyLocate	EasyLocate tool.
EasyLocateInterestPoint	EasyLocateInterestPoint tool.
DeepOCR	A tool for the deep Learning models from EasyDeepOCR.

EDeepOCRModel

enum eVision.EasyDeepLearning.EDeepOCRModel : int

EasyDeepOCR models that are trainable with EDeepOCRTool.

Enumerator
Location	Location model.
Orientation	Orientation model (model used to determine the reading direction).
Recognition	Recognition model (model used to decode the images to a string).
None	No model.

EDLDataAugmentationType

enum eVision.EasyDeepLearning.EDLDataAugmentationType : int

This enum represents how the data augmentation transformation are generated.

Enumerator
Random	Random transformation between the lower and upper limits.
LowerLimit	Transformation using the lower limits.
UpperLimit	Transformation using the upper limits.

EHeatmapColormap

enum eVision.EasyDeepLearning.EHeatmapColormap : int

Color maps for visualizing the heatmap.

Enumerator
YellowToRed	Colormap showing a gradient from yellow to red.
Jet	Colormap showing a gradient going from blue to green to yellow to dark red.
Grayscale	Colormap showing a gradient going from black to white.

EImageAnnotationFormat

enum eVision.EasyDeepLearning.EImageAnnotationFormat : int

Supported formats of annotation files for image file.

Enumerator
None	No annotation.
YOLOTXT	The YOLO annotation format for EasyLocate bounding box. Annotation file: imagefilename.txt. Description of the format: each row of the file defines an object: class_id center_x center_y width height The label list must be correctly set before adding images with YOLO annotations. You can also import a YOLO dataset defined with a YAML file using EDataset.ImportYOLOTXTAnnotations.
PASCALVOCXML	Pascal VOC XML format for EasyLocate bounding box. Annotation file: imagefilename.xml. Description of the format: XML file. See the official Pascal VOC website for a description of the format.

ELocatorCapacity

enum eVision.EasyDeepLearning.ELocatorCapacity : int

The capacity of the locator deep learning network.

A larger capacity means that the underlying neural network is capable of learning more information but it will be slower.

Enumerator
Small	Small capacity.
Normal	Normal capacity.
Large	Large capacity.

ELocatorFeature

enum eVision.EasyDeepLearning.ELocatorFeature : int

Features supported by the ELocator or ELocatorObject.

Enumerator
None	No features.
Position	The locator predicts the position of the object.
Size	The locator predicts the size of the object (width and height).

ESupervisedSegmenterCapacity

enum eVision.EasyDeepLearning.ESupervisedSegmenterCapacity : int

The capacity of the supervised segmentation network.

A larger capacity means that the underlying neural network is capable of learning more information but it will be slower.

Enumerator
Small	Small capacity.
Normal	Normal capacity.
Large	Large capacity.

EUnsupervisedSegmenterCapacity

enum eVision.EasyDeepLearning.EUnsupervisedSegmenterCapacity : int

The capacity of the unsupervised network.

A larger capacity means that the underlying neural network is capable of learning more information but it will be slower.

Enumerator
Small	Small capacity.
Normal	Normal capacity.
Large	Large capacity.