EClassifier allows to train a classifier using an EDataset object and classify new images. More...

Inheritance diagram for eVision.EasyDeepLearning.EClassifier:

[legend]

Collaboration diagram for eVision.EasyDeepLearning.EClassifier:

[legend]

Public Member Functions
	EClassifier ()
	Constructs a EClassifier object.
	EClassifier (eVision.EasyDeepLearning.EClassifier other)
	Copy constructor of a EClassifier object.
override void	SerializeSettings (eVision.ESerializer serializer)
	Serializes the classifier settings.
void	LoadAsPretrained (string path)
	Loads the classifier and use it as a pre-trained model.
void	LoadOnnxModelAsPretrained (string path)
	Loads the ONNX model as a pre-trained model.
void	LoadOnnxModel (string path)
	Loads the ONNX model for inference.
bool	HasPretrainedModel ()
	Whether a pre-trained model for the current configuration can be found.
eVision.EasyDeepLearning.EClassificationMetrics	GetTrainingMetrics (int id)
	Gets the metrics obtained with the training dataset at the given iteration.
eVision.EasyDeepLearning.EClassificationMetrics	GetValidationMetrics (int id)
	Gets the metrics obtained with the validation dataset at the given iteration.
void	Train (eVision.EasyDeepLearning.EDataset dataset)
	Trains the EClassifier with the given dataset for the specified number of iterations.
void	Train (eVision.EasyDeepLearning.EDataset dataset, eVision.EasyDeepLearning.EDataAugmentation dataAugmentation)
	Trains the EClassifier with the given dataset for the specified number of iterations.
void	Train (eVision.EasyDeepLearning.EDataset trainingDataset, eVision.EasyDeepLearning.EDataset validationDataset)
	Trains the EClassifier with the given dataset for the specified number of iterations.
void	Train (eVision.EasyDeepLearning.EDataset trainingDataset, eVision.EasyDeepLearning.EDataset validationDataset, eVision.EasyDeepLearning.EDataAugmentation dataAugmentation)
	Trains the EClassifier with the given dataset for the specified number of iterations.
void	Train (eVision.EasyDeepLearning.EDataset dataset, int iterations)
	Trains the EClassifier with the given dataset for the specified number of iterations.
void	Train (eVision.EasyDeepLearning.EDataset dataset, eVision.EasyDeepLearning.EDataAugmentation dataAugmentation, int iterations)
	Trains the EClassifier with the given dataset for the specified number of iterations.
void	Train (eVision.EasyDeepLearning.EDataset trainingDataset, eVision.EasyDeepLearning.EDataset validationDataset, int iterations)
	Trains the EClassifier with the given dataset for the specified number of iterations.
void	Train (eVision.EasyDeepLearning.EDataset trainingDataset, eVision.EasyDeepLearning.EDataset validationDataset, eVision.EasyDeepLearning.EDataAugmentation dataAugmentation, int iterations)
	Trains the EClassifier with the given dataset for the specified number of iterations.
eVision.EasyDeepLearning.EClassificationResult	Classify (eVision.EBaseROI img)
	Classifies images and returns the complete results as an EClassificationResult object.
eVision.EasyDeepLearning.EClassificationResult	Classify (eVision.EBaseROI img, eVision.ERegion mask)
	Classifies images and returns the complete results as an EClassificationResult object.
eVision.EasyDeepLearning.EClassificationResult[]	Classify (ref eVision.EBaseROI[] imgList, eVision.ERegion mask)
	Classifies images and returns the complete results as an EClassificationResult object.
eVision.EasyDeepLearning.EClassificationResult[]	Classify (ref eVision.EBaseROI[] imgList)
	Classifies images and returns the complete results as an EClassificationResult object.
eVision.EasyDeepLearning.EClassificationResult	Classify (eVision.EBaseROI img, eVision.ERegion mask, eVision.EFrame frame)
	Classifies images and returns the complete results as an EClassificationResult object.
eVision.EasyDeepLearning.EClassificationResult[]	Classify (ref eVision.EBaseROI[] imgList, eVision.ERegion mask, eVision.EFrame frame)
	Classifies images and returns the complete results as an EClassificationResult object.
eVision.EasyDeepLearning.EClassificationResult[]	Classify (ref eVision.EImageBW8[] imgList, eVision.ERegion mask)
	Classifies images and returns the complete results as an EClassificationResult object.
eVision.EasyDeepLearning.EClassificationResult[]	Classify (ref eVision.EImageBW8[] imgList)
	Classifies images and returns the complete results as an EClassificationResult object.
eVision.EasyDeepLearning.EClassificationResult[]	Classify (ref eVision.EImageBW16[] imgList, eVision.ERegion mask)
	Classifies images and returns the complete results as an EClassificationResult object.
eVision.EasyDeepLearning.EClassificationResult[]	Classify (ref eVision.EImageBW16[] imgList)
	Classifies images and returns the complete results as an EClassificationResult object.
eVision.EasyDeepLearning.EClassificationResult[]	Classify (ref eVision.EImageC24[] imgList, eVision.ERegion mask)
	Classifies images and returns the complete results as an EClassificationResult object.
eVision.EasyDeepLearning.EClassificationResult[]	Classify (ref eVision.EImageC24[] imgList)
	Classifies images and returns the complete results as an EClassificationResult object.
eVision.EasyDeepLearning.EClassificationMetrics	Evaluate (eVision.EasyDeepLearning.EDataset dataset)
	Evaluates the EClassifier using the given EDataset.
eVision.EImageBW8	GetHeatMap (eVision.EBaseROI image, string label)
	Gets a heatmap associated with the given label. A heatmap is an image that indicates which pixels in the original image best explain the given label.
eVision.EImageC24	GetColorizedHeatMap (eVision.EBaseROI image, string label)
	Gets a colorized heatmap associated with the given label. A heatmap is an image that indicates which pixels in the original image best explain the given label.
eVision.EImageC24A	GetColorizedHeatmapWithTransparency (eVision.EBaseROI image, string label, float minAlpha, float maxAlpha, eVision.EasyDeepLearning.EHeatmapColormap colormap)
	Colorized heatmap with transparency.
eVision.EImageC24A	GetColorizedHeatmapWithTransparency (eVision.EBaseROI image, string label, float minAlpha, float maxAlpha)
	Colorized heatmap with transparency.
void	LearnOODParameters (eVision.EasyDeepLearning.EDataset trainList, eVision.EasyDeepLearning.EDataset valList)
	Learns OOD parameters using training data. The dataset must correspond to the dataset used to train the classifier.
void	LearnOODParameters (eVision.EBaseROI[] trainList, eVision.EBaseROI[] valList)
	Learns OOD parameters using training data. The dataset must correspond to the dataset used to train the classifier.
void	LearnOODParameters (eVision.EImageBW8[] trainList, eVision.EImageBW8[] valList)
	Learns OOD parameters using training data. The dataset must correspond to the dataset used to train the classifier.
void	LearnOODParameters (eVision.EImageBW16[] trainList, eVision.EImageBW16[] valList)
	Learns OOD parameters using training data. The dataset must correspond to the dataset used to train the classifier.
void	LearnOODParameters (eVision.EImageC24[] trainList, eVision.EImageC24[] valList)
	Learns OOD parameters using training data. The dataset must correspond to the dataset used to train the classifier.
bool	HasOODParameters ()
	Checks whether the OOD parameters have been learned.
Public Member Functions inherited from eVision.EasyDeepLearning.EDeepLearningTool
virtual void	Load (string filePath)
	Loads a deep learning tool. The given ESerializer must have been created for reading.
virtual void	Load (eVision.ESerializer serializer)
	Loads a deep learning tool. The given ESerializer must have been created for reading.
eVision.EasyDeepLearning.EDeepLearningTool	Clone ()
	Clones this tool.
virtual void	Save (string filePath, bool includeTrainingModel)
	Saves the settings, inference and training model of the deep learning tool.
virtual void	Save (string filePath)
	Saves the settings, inference and training model of the deep learning tool.
virtual void	Save (eVision.ESerializer serializer, bool includeTrainingModel)
	Saves the settings, inference and training model of the deep learning tool.
virtual void	Save (eVision.ESerializer serializer)
	Saves the settings, inference and training model of the deep learning tool.
void	SerializeSettings (eVision.ESerializer serializer)
	Serializes the tool settings.
void	SerializeInferenceModel (eVision.ESerializer serializer)
	Serializes the inference model of the tool.
void	SerializeTrainingModel (eVision.ESerializer serializer)
	Serializes the training model of the tool.
void	SaveSettings (string filePath)
	Saves the settings of the tool to a file.
void	SaveInferenceModel (string filePath)
	Saves the tool's inference model to a file.
void	SaveTrainingModel (string filePath)
	Saves the tool's training model to a file.
void	LoadSettings (string filePath)
	Loads the settings of a tool.
void	LoadInferenceModel (string filePath)
	Loads the tool's inference model. The inference model is required to apply the tool on new images.
void	LoadTrainingModel (string filePath)
	Loads the tool's training model. The training model is required to continue a training.
void	UnloadInferenceModel ()
	Unloads the inference model.
void	UnloadTrainingModel ()
	Unloads the training model.
void	UnloadModels ()
	Unloads the inference and training model.
bool	HasInferenceModel ()
	Whether the inference model is loaded or created when the tool is not yet trained.
bool	HasTrainingModel ()
	Whether the training model is loaded or created when the tool is not yet trained.
bool	IsTrained ()
	Tells whether the deep learning tool has been trained.
string	GetAvailableDeviceName (int i)
	Name of the i-th device of the current engine.
eVision.EasyDeepLearning.EDeepLearningDevice	GetAvailableDevice (int i)
	i-th EDeepLearningDevice of the current engine.
void	SetActiveDevice (eVision.EasyDeepLearning.EDeepLearningDevice device)
	Sets the device with which to run or train the model.
eVision.EasyDeepLearning.EDeepLearningDevice	GetActiveDevice (int i)
	Gets the active device. By default, it returns the first active device.
eVision.EasyDeepLearning.EDeepLearningDevice	GetActiveDevice ()
	Gets the active device. By default, it returns the first active device.
bool	HasEngineSetupData ()
	Whether the inference model has engine setup data.
void	ClearEngineSetupData ()
	Clears the engine data saved in the inference model. Use this method when your engine fails.
bool	IsTraining ()
	Indicates whether the object is currently training.
int	WaitForTrainingCompletion (int timeout)
	Waits until the training is complete or the timeout is expired. A call to this method will block the calling thread for the shortest time between the timeout and the time it takes for the training to complete.
int	WaitForTrainingCompletion ()
	Waits until the training is complete or the timeout is expired. A call to this method will block the calling thread for the shortest time between the timeout and the time it takes for the training to complete.
int	WaitForIterationCompletion ()
	Waits until an iteration is complete. A call to this method will block the calling thread until a training iteration in the training thread is finished. This method returns the number of trained iterations.
int	StopTraining (bool wait)
	Stops training and returns the last completed iteration. If the parameter 'wait' is set to true, the method will wait for the training thread to completely stop. Otherwise, the method will return immediately.
void	SetLearningRate (eVision.EasyDeepLearning.EDeepLearningLearningRateType type, float[] parameters)
	Specifies how the value of the learning rate and how it evolves during training.
void	InitializeInference (int width, int height, eVision.EImageType imageType, int numImages)
	Initializes the neural network for inference (i.e. making predictions with the neural network).
void	InitializeInference (eVision.EBaseROI img)
	Initializes the neural network for inference (i.e. making predictions with the neural network).
void	InitializeInference (ref eVision.EBaseROI[] imgList)
	Initializes the neural network for inference (i.e. making predictions with the neural network).
void	InitializeInference (ref eVision.EImageBW8[] imgList)
	Initializes the neural network for inference (i.e. making predictions with the neural network).
void	InitializeInference (ref eVision.EImageBW16[] imgList)
	Initializes the neural network for inference (i.e. making predictions with the neural network).
void	InitializeInference (ref eVision.EImageC24[] imgList)
	Initializes the neural network for inference (i.e. making predictions with the neural network).
string	GetLabel (uint index)
	Gets the label from its index. If the tool is not trained, the method may throw an exception.
virtual void	SetLabel (uint index, string label)
	Changes a label predicted by the tool.
void	SetLabelWeight (uint index, float weight)
	Sets the label weight. If the tool is not trained, the label weights are not defined. After training, they are set to the value of the training dataset label weights.
float	GetLabelWeight (uint index)
	Gets the label weight. If the tool is not trained, the label weights are not defined. After training, they are set to the value of the training dataset label weights.
void	SetLabelColor (uint index, eVision.ERGBColor c)
	Changes a label color.
eVision.ERGBColor	GetLabelColor (uint index)
	Label color. The label colors affect how the results will be displayed.
void	SetLabelOpacity (uint index, byte opacity)
	Changes a label opacity (between 0 for fully transparent and 255 for opaque color).
byte	GetLabelOpacity (uint index)
	Gets the label opacity (between 0 for fully transparent and 255 for opaque color). The label opacities affect with the label colors how the results will be displayed.
int	GetOptimalNumImagesForBatchSize (eVision.EBaseROI sampleImage)
	Optimal number of images to process together for inference given the batch size and EDeepLearningTool.GetNumPatchesForImage.
int	GetOptimalNumImagesForBatchSize (int imageWidth, int imageHeight)
	Optimal number of images to process together for inference given the batch size and EDeepLearningTool.GetNumPatchesForImage.
int	GetNumPatchesForImage (eVision.EBaseROI sampleImage)
	Number of patches that will be extracted from an input image to perform inference.
virtual int	GetNumPatchesForImage (int imageWidth, int imageHeight)
	Number of patches that will be extracted from an input image to perform inference.
virtual void	Clean ()
	Reset the engine and clean CPU and GPU memory allocated for running the model.

Static Public Member Functions
static string	GetAvailableModelTypes (int i)
	Get the i-th available models to be used with EClassifier.ModelType.
Static Public Member Functions inherited from eVision.EasyDeepLearning.EDeepLearningTool
static eVision.EasyDeepLearning.EDeepLearningTool	Create (string path)
	Factory method to open a deep learning tool.
static eVision.EasyDeepLearning.EDeepLearningTool	Create (eVision.ESerializer serializer)
	Factory method to open a deep learning tool.
static eVision.EasyDeepLearning.EDeepLearningTool	Create (eVision.EasyDeepLearning.EDeepLearningToolType type)
	Factory method to open a deep learning tool.
static string	GetAvailableEngineName (int id)
	Gets the name of the engine given its id.
static eVision.EasyDeepLearning.EDeepLearningDevice[]	GetAvailableDevicesForEngine (string engine)
	Available devices for the specified engine.

Properties
override eVision.EasyDeepLearning.EDeepLearningToolType	ToolType `[get]`
	Type of the deep learning tool.
bool	UsePretrainedModel `[get, set]`
	Whether to use a pre-trained model when training.
uint	Width `[get, set]`
	Width for input images of the classifier. By default, this value will be set to 224 if it wasn't explicitly set before training.
uint	Height `[get, set]`
	Height for input images of the classifier. By default, this value will be set to 224 if it wasn't explicitly set before training.
uint	Channels `[get, set]`
	Number of channels for input images of the classifier. The number of channels can be either 1 (monochrome image) or 3 (RGB image). By default, this value will be set from the format of the first image added to the training dataset.
bool	EnableAutomaticImageReformat `[get, set]`
	Enable automatic image reformat (true by default).
bool	EnableHistogramEqualization `[get, set]`
	Enable histogram equalization of all images passing through the classifier. (false by default).
uint	MinimumWidth `[get]`
	Minimum width for input images of the classifier. This value is equal to 1 for standard networks.
uint	MinimumHeight `[get]`
	Minimum height for input images of the classifier. This value is equal to 1 for standard networks.
eVision.EasyDeepLearning.EClassifierCapacity	Capacity `[get, set]`
	Capacity of classifier to use. Be aware that changing the classifier capacity will delete the effect of any previous training. The capacity will be translated to the corresponding type (Small, Normal, or Large).
string	ModelType `[get, set]`
	Model type for training. Default: "Normal".
float	OODThreshold `[get, set]`
	Out-Of-Distribution (OOD) threshold below which an image is considered to be OOD.
bool	ComputeHeatmapWithResult `[get, set]`
	Whether to always compute the heatmap along the result when applying the tool. If true, the heatmap is available through EClassificationResult.Heatmap.
bool	ComputeOODWithResult `[get, set]`
	Whether to compute the OOD parameters while training. Also determines whether the OOD score is computed alongside the result when applying the tool. If true, the OOD score is available through EClassificationResult.OODScore.
static int	NumAvailableModelTypes `[get]`
	Number of available models for training an EasyClassify tool.
Properties inherited from eVision.EasyDeepLearning.EDeepLearningTool
eVision.EasyDeepLearning.EDeepLearningToolType	ToolType `[get]`
	Type of the deep learning tool.
bool	EnableGPU `[get, set]`
	Enable the use of a GPU for training and inference of the deep learning tool. If the active engine is not the default engine, EDeepLearningTool.EnableGPU will be false even if the active device has a GPU device type. Setting EDeepLearningTool.EnableGPU to any value when the active engine is not the default one will throw an exception.
uint[]	GPUIndexes `[get, set]`
	Indexes of the GPUs to use for computations.
uint	NumGPUs `[get]`
	Number of detected NVIDIA GPU.
string	Engine `[get, set]`
	Sets the engine with which to execute or train the neural network.
eVision.EasyDeepLearning.EDeepLearningInferencePrecision	InferencePrecision `[get, set]`
	Inference precision.
int	NumAvailableDevices `[get]`
	Available devices for the current EDeepLearningTool.Engine.
eVision.EasyDeepLearning.EDeepLearningDevice[]	AvailableDevices `[get]`
	Get all available device for the current engine.
int	NumActiveDevices `[get]`
	Gets the number of devices with which to run or train the model.
string	ActiveDeviceByName `[set]`
	Sets the device with which to run the model using its name.
int	ActiveDeviceById `[set]`
	Sets the device with which to run the model using its index in the list returned by EDeepLearningTool.AvailableDevices.
eVision.EasyDeepLearning.EDeepLearningDevice[]	ActiveDevices `[get, set]`
	Devices with which to run or train the model.
int[]	ActiveDevicesById `[set]`
	Sets the devices with which to run the model using their index in the list returned by EDeepLearningTool.AvailableDevices.
eVision.EasyDeepLearning.EDeepLearningExecutionSettings	ExecutionSettings `[get, set]`
	Execution settings of the tool (engine, device(s), inference precision, batch size settings).
eVision.EasyDeepLearning.EDeepLearningDevice	EngineSetupDataDevice `[get]`
	Device corresponding to the engine data contained in the tool.
eVision.EasyDeepLearning.EDeepLearningInferencePrecision	EngineSetupDataInferencePrecision `[get]`
	Inference precision corresponding to the engine data contained in the tool.
int	EngineSetupDataBatchSize `[get]`
	Batch size corresponding to the engine data contained in the tool.
bool	OptimizeBatchSize `[get, set]`
	Indicates whether to optimize the batch size (see EDeepLearningTool.BatchSize) to maximize the training and inference speed according to the engine, the device and the available memory.
int	BatchSizeForMaximumInferenceSpeed `[get]`
	Computes the batch size that will maximize the inference speed on NVIDIA GPU. For all other devices, it will return 1.
int	BatchSize `[get, set]`
	Batch size. The batch size is the number of images that are processed together during training and batch inference.
float	CurrentTrainingProgression `[get]`
	Current training progression as a percentage (between 0 and 1).
uint	CurrentTrainingFinishedIterations `[get]`
	Number of iterations that are already finished in the current training.
uint	CurrentTrainingNumIterations `[get]`
	Total number of training iterations that will be performed during the current training of the deep learning tool.
uint	NumTrainedIterations `[get]`
	Number of iterations that were performed to train this deep learning tool.
int	BestIteration `[get]`
	Iteration at which the minimum validation error was reached. After training, the tool is in the state it was at this best iteration.
ulong	ImageCacheSize `[get, set]`
	Size in byte of the image cache. The cache is used during training to store reformatted and normalized images. A correctly sized cache can reduce the hard drive accesses and the preprocessing time for each image.
string	Path `[get]`
	Path of the tool when loaded or saved using EDeepLearningTool.Save or EDeepLearningTool.Load.
string	InferenceModelPath `[get]`
	Path of the inference model when loaded or saved using EDeepLearningTool.SaveInferenceModel, EDeepLearningTool.LoadInferenceModel or EDeepLearningTool.SerializeInferenceModel.
string	TrainingModelPath `[get]`
	Path of the training model when loaded or saved using EDeepLearningTool.SaveTrainingModel, EDeepLearningTool.LoadTrainingModel or EDeepLearningTool.SerializeTrainingModel.
string	SettingsPath `[get]`
	Path of the settings when loaded or saved using EDeepLearningTool.SaveSettings, EDeepLearningTool.LoadSettings or EDeepLearningTool.SerializeSettings.
int	AutoSavePeriod `[get, set]`
	Auto-save period. The auto-save period is the number of training iterations between each automatic save of the model.
eVision.EasyDeepLearning.EDeepLearningLearningRateType	LearningRateType `[get]`
	Learning rate type.
float[]	LearningRateParameters `[get]`
	Parameters of the learning rate.
bool	EnableDeterministicTraining `[get, set]`
	Whether to use deterministic training algorithm that allows to repeatable training results. Default value: false.
int	DeterministicTrainingRandomSeed `[get, set]`
	Random seed to use when EDeepLearningTool.EnableDeterministicTraining is set to true. The default value is 42.
uint	NumLabels `[get]`
	Number of labels of this tool. If the tool is not trained, the method will throw an exception.
static int	NumAvailableEngines `[get]`
	Number of available engines.
static string[]	AvailableEngines `[get]`
	Available engines.

Detailed Description

EClassifier allows to train a classifier using an EDataset object and classify new images.

As required by Deep Learning techniques, the input image of EClassifier must be of the same format (width, height, number of channels). By default, this format will be 224x224xC (which is the format used for pre-training our networks) with C being the number of channel of the first image of the dataset. The format can also be specified by the EClassifier.Width, EClassifier.Height and EClassifier.Channels unless its width and height is smaller than the minimum width and height supported by the classifier. In this case, the input resolution will be the minimum resolution supported by the classifier. (See EClassifier.MinimumWidth and EClassifier.MinimumHeight). methods.

By default, images that don't satisfy the image format of the classifier are automatically reformatted. This behavior can be controlled through the EClassifier.EnableAutomaticImageReformat method. When the automatic image reformatting is disabled, training or classifying an image that doesn't satisfy the input image format will result in an exception.

Once trained, the input image format cannot be changed.

License(s): EasyClassify or EasyClassifyInference

Constructor & Destructor Documentation

◆ EClassifier() [1/2]

eVision.EasyDeepLearning.EClassifier.EClassifier ( )

Constructs a EClassifier object.

◆ EClassifier() [2/2]

eVision.EasyDeepLearning.EClassifier.EClassifier ( eVision.EasyDeepLearning.EClassifier other )

Copy constructor of a EClassifier object.

Parameters

other Reference to the EClassifier object that should be copied

Member Function Documentation

◆ Classify() [1/12]

eVision.EasyDeepLearning.EClassificationResult eVision.EasyDeepLearning.EClassifier.Classify ( eVision.EBaseROI img )

Classifies images and returns the complete results as an EClassificationResult object.

The method throws an exception if the input image does not fulfill the input specification.

Parameters

img	Image to classify

Classifying a set of images is usually faster than classifying each image sequentially.

To maximize the classification speed on a GPU, EDeepLearningTool.BatchSize and the size of the set of input images must be equal to the value returned by EDeepLearningTool.BatchSizeForMaximumInferenceSpeed.

◆ Classify() [2/12]

eVision.EasyDeepLearning.EClassificationResult eVision.EasyDeepLearning.EClassifier.Classify	(	eVision.EBaseROI	img,
		eVision.ERegion	mask )

Classifies images and returns the complete results as an EClassificationResult object.

The method throws an exception if the input image does not fulfill the input specification.

Parameters

img	Image to classify
mask	Mask of image to classify

Classifying a set of images is usually faster than classifying each image sequentially.

To maximize the classification speed on a GPU, EDeepLearningTool.BatchSize and the size of the set of input images must be equal to the value returned by EDeepLearningTool.BatchSizeForMaximumInferenceSpeed.

◆ Classify() [3/12]

eVision.EasyDeepLearning.EClassificationResult eVision.EasyDeepLearning.EClassifier.Classify	(	eVision.EBaseROI	img,
		eVision.ERegion	mask,
		eVision.EFrame	frame )

Classifies images and returns the complete results as an EClassificationResult object.

The method throws an exception if the input image does not fulfill the input specification.

Parameters

img	Image to classify
mask	Mask of image to classify
frame	Frame associated with the mask.

Classifying a set of images is usually faster than classifying each image sequentially.

To maximize the classification speed on a GPU, EDeepLearningTool.BatchSize and the size of the set of input images must be equal to the value returned by EDeepLearningTool.BatchSizeForMaximumInferenceSpeed.

◆ Classify() [4/12]

eVision.EasyDeepLearning.EClassificationResult[] eVision.EasyDeepLearning.EClassifier.Classify ( ref eVision.EBaseROI[] imgList )

Classifies images and returns the complete results as an EClassificationResult object.

The method throws an exception if the input image does not fulfill the input specification.

Parameters

imgList Vector of images to classify

Classifying a set of images is usually faster than classifying each image sequentially.

To maximize the classification speed on a GPU, EDeepLearningTool.BatchSize and the size of the set of input images must be equal to the value returned by EDeepLearningTool.BatchSizeForMaximumInferenceSpeed.

◆ Classify() [5/12]

eVision.EasyDeepLearning.EClassificationResult[] eVision.EasyDeepLearning.EClassifier.Classify	(	ref eVision.EBaseROI[]	imgList,
		eVision.ERegion	mask )

Classifies images and returns the complete results as an EClassificationResult object.

The method throws an exception if the input image does not fulfill the input specification.

Parameters

imgList	Vector of images to classify
mask	Mask of image to classify

Classifying a set of images is usually faster than classifying each image sequentially.

To maximize the classification speed on a GPU, EDeepLearningTool.BatchSize and the size of the set of input images must be equal to the value returned by EDeepLearningTool.BatchSizeForMaximumInferenceSpeed.

◆ Classify() [6/12]

eVision.EasyDeepLearning.EClassificationResult[] eVision.EasyDeepLearning.EClassifier.Classify	(	ref eVision.EBaseROI[]	imgList,
		eVision.ERegion	mask,
		eVision.EFrame	frame )

Classifies images and returns the complete results as an EClassificationResult object.

The method throws an exception if the input image does not fulfill the input specification.

Parameters

imgList	Vector of images to classify
mask	Mask of image to classify
frame	Frame associated with the mask.

Classifying a set of images is usually faster than classifying each image sequentially.

To maximize the classification speed on a GPU, EDeepLearningTool.BatchSize and the size of the set of input images must be equal to the value returned by EDeepLearningTool.BatchSizeForMaximumInferenceSpeed.

◆ Classify() [7/12]

eVision.EasyDeepLearning.EClassificationResult[] eVision.EasyDeepLearning.EClassifier.Classify ( ref eVision.EImageBW16[] imgList )

Classifies images and returns the complete results as an EClassificationResult object.

The method throws an exception if the input image does not fulfill the input specification.

Parameters

imgList Vector of images to classify

Classifying a set of images is usually faster than classifying each image sequentially.

To maximize the classification speed on a GPU, EDeepLearningTool.BatchSize and the size of the set of input images must be equal to the value returned by EDeepLearningTool.BatchSizeForMaximumInferenceSpeed.

◆ Classify() [8/12]

eVision.EasyDeepLearning.EClassificationResult[] eVision.EasyDeepLearning.EClassifier.Classify	(	ref eVision.EImageBW16[]	imgList,
		eVision.ERegion	mask )

Classifies images and returns the complete results as an EClassificationResult object.

The method throws an exception if the input image does not fulfill the input specification.

Parameters

imgList	Vector of images to classify
mask	Mask of image to classify

Classifying a set of images is usually faster than classifying each image sequentially.

To maximize the classification speed on a GPU, EDeepLearningTool.BatchSize and the size of the set of input images must be equal to the value returned by EDeepLearningTool.BatchSizeForMaximumInferenceSpeed.

◆ Classify() [9/12]

eVision.EasyDeepLearning.EClassificationResult[] eVision.EasyDeepLearning.EClassifier.Classify ( ref eVision.EImageBW8[] imgList )

Classifies images and returns the complete results as an EClassificationResult object.

The method throws an exception if the input image does not fulfill the input specification.

Parameters

imgList Vector of images to classify

Classifying a set of images is usually faster than classifying each image sequentially.

To maximize the classification speed on a GPU, EDeepLearningTool.BatchSize and the size of the set of input images must be equal to the value returned by EDeepLearningTool.BatchSizeForMaximumInferenceSpeed.

◆ Classify() [10/12]

eVision.EasyDeepLearning.EClassificationResult[] eVision.EasyDeepLearning.EClassifier.Classify	(	ref eVision.EImageBW8[]	imgList,
		eVision.ERegion	mask )

Classifies images and returns the complete results as an EClassificationResult object.

The method throws an exception if the input image does not fulfill the input specification.

Parameters

imgList	Vector of images to classify
mask	Mask of image to classify

Classifying a set of images is usually faster than classifying each image sequentially.

To maximize the classification speed on a GPU, EDeepLearningTool.BatchSize and the size of the set of input images must be equal to the value returned by EDeepLearningTool.BatchSizeForMaximumInferenceSpeed.

◆ Classify() [11/12]

eVision.EasyDeepLearning.EClassificationResult[] eVision.EasyDeepLearning.EClassifier.Classify ( ref eVision.EImageC24[] imgList )

Classifies images and returns the complete results as an EClassificationResult object.

The method throws an exception if the input image does not fulfill the input specification.

Parameters

imgList Vector of images to classify

Classifying a set of images is usually faster than classifying each image sequentially.

To maximize the classification speed on a GPU, EDeepLearningTool.BatchSize and the size of the set of input images must be equal to the value returned by EDeepLearningTool.BatchSizeForMaximumInferenceSpeed.

◆ Classify() [12/12]

eVision.EasyDeepLearning.EClassificationResult[] eVision.EasyDeepLearning.EClassifier.Classify	(	ref eVision.EImageC24[]	imgList,
		eVision.ERegion	mask )

Classifies images and returns the complete results as an EClassificationResult object.

The method throws an exception if the input image does not fulfill the input specification.

Parameters

imgList	Vector of images to classify
mask	Mask of image to classify

Classifying a set of images is usually faster than classifying each image sequentially.

To maximize the classification speed on a GPU, EDeepLearningTool.BatchSize and the size of the set of input images must be equal to the value returned by EDeepLearningTool.BatchSizeForMaximumInferenceSpeed.

◆ Evaluate()

eVision.EasyDeepLearning.EClassificationMetrics eVision.EasyDeepLearning.EClassifier.Evaluate ( eVision.EasyDeepLearning.EDataset dataset )

Evaluates the EClassifier using the given EDataset.

Parameters

dataset EDataset with which to evaluate the classifier

This method computes various metrics (see EClassificationMetrics on the given dataset.

The method ignores the label weights and the data augmentation settings of the given dataset. However, the label weights can be taken into consideration in the returned EClassificationMetrics object.

License(s): EasyClassify

◆ GetAvailableModelTypes()

string eVision.EasyDeepLearning.EClassifier.GetAvailableModelTypes ( int i )

static

Get the i-th available models to be used with EClassifier.ModelType.

Parameters

i	Index of the model between 0 and EClassifier.NumAvailableModelTypes - 1.

◆ GetColorizedHeatMap()

eVision.EImageC24 eVision.EasyDeepLearning.EClassifier.GetColorizedHeatMap	(	eVision.EBaseROI	image,
		string	label )

Gets a colorized heatmap associated with the given label. A heatmap is an image that indicates which pixels in the original image best explain the given label.

Parameters

image	A reference to the image we want to generate the heatmap from.
label	A reference to the string representing the label we want to explain for the given image.

◆ GetColorizedHeatmapWithTransparency() [1/2]

eVision.EImageC24A eVision.EasyDeepLearning.EClassifier.GetColorizedHeatmapWithTransparency	(	eVision.EBaseROI	image,
		string	label,
		float	minAlpha,
		float	maxAlpha )

Colorized heatmap with transparency.

When the minimum and maximum alpha values are different from each other, the transparency value for a pixel will depend on the value of the heatmap at that pixel. The range of heatmap values [0, 255] will be mapped to the range [minAlpha, maxAlpha].

Parameters

image	A reference to the image we want to generate the heatmap from.
label	A reference to the string representing the label we want to explain for the given image.
minAlpha	Minimum transparency value.
maxAlpha	Maximum transparency value.

◆ GetColorizedHeatmapWithTransparency() [2/2]

eVision.EImageC24A eVision.EasyDeepLearning.EClassifier.GetColorizedHeatmapWithTransparency	(	eVision.EBaseROI	image,
		string	label,
		float	minAlpha,
		float	maxAlpha,
		eVision.EasyDeepLearning.EHeatmapColormap	colormap )

Colorized heatmap with transparency.

When the minimum and maximum alpha values are different from each other, the transparency value for a pixel will depend on the value of the heatmap at that pixel. The range of heatmap values [0, 255] will be mapped to the range [minAlpha, maxAlpha].

Parameters

image	A reference to the image we want to generate the heatmap from.
label	A reference to the string representing the label we want to explain for the given image.
minAlpha	Minimum transparency value.
maxAlpha	Maximum transparency value.
colormap	Color map

◆ GetHeatMap()

eVision.EImageBW8 eVision.EasyDeepLearning.EClassifier.GetHeatMap	(	eVision.EBaseROI	image,
		string	label )

Gets a heatmap associated with the given label. A heatmap is an image that indicates which pixels in the original image best explain the given label.

Parameters

image	A reference to the image we want to generate the heatmap from.
label	A reference to the string representing the label we want to explain for the given image.

◆ GetTrainingMetrics()

eVision.EasyDeepLearning.EClassificationMetrics eVision.EasyDeepLearning.EClassifier.GetTrainingMetrics ( int id )

Gets the metrics obtained with the training dataset at the given iteration.

The iterations are indexed between 0 and EDeepLearningTool.NumTrainedIterations - 1.

Parameters

id	The iteration index

◆ GetValidationMetrics()

eVision.EasyDeepLearning.EClassificationMetrics eVision.EasyDeepLearning.EClassifier.GetValidationMetrics ( int id )

Gets the metrics obtained with the validation dataset at the given iteration.

The iterations are indexed between 0 and EDeepLearningTool.NumTrainedIterations - 1.

Parameters

id	The iteration index

◆ HasOODParameters()

bool eVision.EasyDeepLearning.EClassifier.HasOODParameters ( )

Checks whether the OOD parameters have been learned.

◆ HasPretrainedModel()

bool eVision.EasyDeepLearning.EClassifier.HasPretrainedModel ( )

Whether a pre-trained model for the current configuration can be found.

◆ LearnOODParameters() [1/5]

void eVision.EasyDeepLearning.EClassifier.LearnOODParameters	(	eVision.EasyDeepLearning.EDataset	trainList,
		eVision.EasyDeepLearning.EDataset	valList )

Learns OOD parameters using training data. The dataset must correspond to the dataset used to train the classifier.

Parameters

trainList	List of images used for training the EClassifier.
valList	List of images used as validation during the training of the EClassifier.

◆ LearnOODParameters() [2/5]

void eVision.EasyDeepLearning.EClassifier.LearnOODParameters	(	eVision.EBaseROI[]	trainList,
		eVision.EBaseROI[]	valList )

Learns OOD parameters using training data. The dataset must correspond to the dataset used to train the classifier.

Parameters

trainList	List of images used for training the EClassifier.
valList	List of images used as validation during the training of the EClassifier.

◆ LearnOODParameters() [3/5]

void eVision.EasyDeepLearning.EClassifier.LearnOODParameters	(	eVision.EImageBW16[]	trainList,
		eVision.EImageBW16[]	valList )

Learns OOD parameters using training data. The dataset must correspond to the dataset used to train the classifier.

Parameters

trainList	List of images used for training the EClassifier.
valList	List of images used as validation during the training of the EClassifier.

◆ LearnOODParameters() [4/5]

void eVision.EasyDeepLearning.EClassifier.LearnOODParameters	(	eVision.EImageBW8[]	trainList,
		eVision.EImageBW8[]	valList )

Learns OOD parameters using training data. The dataset must correspond to the dataset used to train the classifier.

Parameters

trainList	List of images used for training the EClassifier.
valList	List of images used as validation during the training of the EClassifier.

◆ LearnOODParameters() [5/5]

void eVision.EasyDeepLearning.EClassifier.LearnOODParameters	(	eVision.EImageC24[]	trainList,
		eVision.EImageC24[]	valList )

Learns OOD parameters using training data. The dataset must correspond to the dataset used to train the classifier.

Parameters

trainList	List of images used for training the EClassifier.
valList	List of images used as validation during the training of the EClassifier.

◆ LoadAsPretrained()

void eVision.EasyDeepLearning.EClassifier.LoadAsPretrained ( string path )

Loads the classifier and use it as a pre-trained model.

Parameters

path	Path to the pre-trained model.

◆ LoadOnnxModel()

void eVision.EasyDeepLearning.EClassifier.LoadOnnxModel ( string path )

Loads the ONNX model for inference.

The methods will automatically create labels according to the number of classes in the ONNX model. The labels are named according to their position, i.e. "0", "1", etc. You can change the labels afterward using the method EDeepLearningTool.SetLabel.

◆ LoadOnnxModelAsPretrained()

void eVision.EasyDeepLearning.EClassifier.LoadOnnxModelAsPretrained ( string path )

Loads the ONNX model as a pre-trained model.

This method loads the architecture of the neural network as well as the trained weights in the model but discards the weights and parameters of some of the last layers to allow retraining on a new dataset with a different number of labels.

◆ SerializeSettings()

override void eVision.EasyDeepLearning.EClassifier.SerializeSettings ( eVision.ESerializer serializer )

Serializes the classifier settings.

Parameters

serializer Pointer to ESerializer

◆ Train() [1/8]

void eVision.EasyDeepLearning.EClassifier.Train ( eVision.EasyDeepLearning.EDataset dataset )

Trains the EClassifier with the given dataset for the specified number of iterations.

At the end of the training, the deep learning tool is in the state it was at the iteration that gave the minimum validation error. See EDeepLearningTool.BestIteration.

If no iterations is given, the training will stop automatically.

Parameters

dataset EDataset with which to train and validate the deep learning tool

License(s): EasyClassify

◆ Train() [2/8]

void eVision.EasyDeepLearning.EClassifier.Train	(	eVision.EasyDeepLearning.EDataset	dataset,
		eVision.EasyDeepLearning.EDataAugmentation	dataAugmentation )

Trains the EClassifier with the given dataset for the specified number of iterations.

At the end of the training, the deep learning tool is in the state it was at the iteration that gave the minimum validation error. See EDeepLearningTool.BestIteration.

If no iterations is given, the training will stop automatically.

Parameters

dataset	EDataset with which to train and validate the deep learning tool
dataAugmentation	Data augmentation to use during training

License(s): EasyClassify

◆ Train() [3/8]

void eVision.EasyDeepLearning.EClassifier.Train	(	eVision.EasyDeepLearning.EDataset	dataset,
		eVision.EasyDeepLearning.EDataAugmentation	dataAugmentation,
		int	iterations )

Trains the EClassifier with the given dataset for the specified number of iterations.

At the end of the training, the deep learning tool is in the state it was at the iteration that gave the minimum validation error. See EDeepLearningTool.BestIteration.

If no iterations is given, the training will stop automatically.

Parameters

dataset	EDataset with which to train and validate the deep learning tool
dataAugmentation	Data augmentation to use during training
iterations	Number of iterations for training.

License(s): EasyClassify

◆ Train() [4/8]

void eVision.EasyDeepLearning.EClassifier.Train	(	eVision.EasyDeepLearning.EDataset	dataset,
		int	iterations )

Trains the EClassifier with the given dataset for the specified number of iterations.

At the end of the training, the deep learning tool is in the state it was at the iteration that gave the minimum validation error. See EDeepLearningTool.BestIteration.

If no iterations is given, the training will stop automatically.

Parameters

dataset	EDataset with which to train and validate the deep learning tool
iterations	Number of iterations for training.

License(s): EasyClassify

◆ Train() [5/8]

void eVision.EasyDeepLearning.EClassifier.Train	(	eVision.EasyDeepLearning.EDataset	trainingDataset,
		eVision.EasyDeepLearning.EDataset	validationDataset )

Trains the EClassifier with the given dataset for the specified number of iterations.

At the end of the training, the deep learning tool is in the state it was at the iteration that gave the minimum validation error. See EDeepLearningTool.BestIteration.

If no iterations is given, the training will stop automatically.

Parameters

trainingDataset	EDataset with which to train the deep learning tool
validationDataset	EDataset with which to validate the deep learning tool

License(s): EasyClassify

◆ Train() [6/8]

void eVision.EasyDeepLearning.EClassifier.Train	(	eVision.EasyDeepLearning.EDataset	trainingDataset,
		eVision.EasyDeepLearning.EDataset	validationDataset,
		eVision.EasyDeepLearning.EDataAugmentation	dataAugmentation )

Trains the EClassifier with the given dataset for the specified number of iterations.

At the end of the training, the deep learning tool is in the state it was at the iteration that gave the minimum validation error. See EDeepLearningTool.BestIteration.

If no iterations is given, the training will stop automatically.

Parameters

trainingDataset	EDataset with which to train the deep learning tool
validationDataset	EDataset with which to validate the deep learning tool
dataAugmentation	Data augmentation to use during training

License(s): EasyClassify

◆ Train() [7/8]

void eVision.EasyDeepLearning.EClassifier.Train	(	eVision.EasyDeepLearning.EDataset	trainingDataset,
		eVision.EasyDeepLearning.EDataset	validationDataset,
		eVision.EasyDeepLearning.EDataAugmentation	dataAugmentation,
		int	iterations )

Trains the EClassifier with the given dataset for the specified number of iterations.

At the end of the training, the deep learning tool is in the state it was at the iteration that gave the minimum validation error. See EDeepLearningTool.BestIteration.

If no iterations is given, the training will stop automatically.

Parameters

trainingDataset	EDataset with which to train the deep learning tool
validationDataset	EDataset with which to validate the deep learning tool
dataAugmentation	Data augmentation to use during training
iterations	Number of iterations for training.

License(s): EasyClassify

◆ Train() [8/8]

void eVision.EasyDeepLearning.EClassifier.Train	(	eVision.EasyDeepLearning.EDataset	trainingDataset,
		eVision.EasyDeepLearning.EDataset	validationDataset,
		int	iterations )

Trains the EClassifier with the given dataset for the specified number of iterations.

At the end of the training, the deep learning tool is in the state it was at the iteration that gave the minimum validation error. See EDeepLearningTool.BestIteration.

If no iterations is given, the training will stop automatically.

Parameters

trainingDataset	EDataset with which to train the deep learning tool
validationDataset	EDataset with which to validate the deep learning tool
iterations	Number of iterations for training.

License(s): EasyClassify

Property Documentation

◆ Capacity

eVision.EasyDeepLearning.EClassifierCapacity eVision.EasyDeepLearning.EClassifier.Capacity

getset

Capacity of classifier to use. Be aware that changing the classifier capacity will delete the effect of any previous training. The capacity will be translated to the corresponding type (Small, Normal, or Large).

Deprecated: Deprecated in version 22.12. Scheduled to be removed in version 27.6.

◆ Channels

uint eVision.EasyDeepLearning.EClassifier.Channels

getset

Number of channels for input images of the classifier. The number of channels can be either 1 (monochrome image) or 3 (RGB image). By default, this value will be set from the format of the first image added to the training dataset.

If the classifier is not trained or the value was not explicitly set, its value will be 0.

◆ ComputeHeatmapWithResult

bool eVision.EasyDeepLearning.EClassifier.ComputeHeatmapWithResult

getset

Whether to always compute the heatmap along the result when applying the tool. If true, the heatmap is available through EClassificationResult.Heatmap.

◆ ComputeOODWithResult

bool eVision.EasyDeepLearning.EClassifier.ComputeOODWithResult

getset

Whether to compute the OOD parameters while training. Also determines whether the OOD score is computed alongside the result when applying the tool. If true, the OOD score is available through EClassificationResult.OODScore.

◆ EnableAutomaticImageReformat

bool eVision.EasyDeepLearning.EClassifier.EnableAutomaticImageReformat

getset

Enable automatic image reformat (true by default).

◆ EnableHistogramEqualization

bool eVision.EasyDeepLearning.EClassifier.EnableHistogramEqualization

getset

Enable histogram equalization of all images passing through the classifier. (false by default).

◆ Height

uint eVision.EasyDeepLearning.EClassifier.Height

getset

Height for input images of the classifier. By default, this value will be set to 224 if it wasn't explicitly set before training.

If the classifier is not trained or the value was not explicitly set, its value will be 0.

◆ MinimumHeight

uint eVision.EasyDeepLearning.EClassifier.MinimumHeight

get

Minimum height for input images of the classifier. This value is equal to 1 for standard networks.

◆ MinimumWidth

uint eVision.EasyDeepLearning.EClassifier.MinimumWidth

get

Minimum width for input images of the classifier. This value is equal to 1 for standard networks.

◆ ModelType

string eVision.EasyDeepLearning.EClassifier.ModelType

getset

Model type for training. Default: "Normal".

The list of available models is available through EClassifier.GetAvailableModelTypes.

◆ NumAvailableModelTypes

int eVision.EasyDeepLearning.EClassifier.NumAvailableModelTypes

staticget

Number of available models for training an EasyClassify tool.

◆ OODThreshold

float eVision.EasyDeepLearning.EClassifier.OODThreshold

getset

Out-Of-Distribution (OOD) threshold below which an image is considered to be OOD.

If training has been carried out with EClassifier.ComputeOODWithResult set to True, the parameters have been automatically initialized. Otherwise, call EClassifier.LearnOODParameters with the training data before calling this method.

The OOD threshold is automatically initialized during learning. If needed, the parameter can also be changed manually.

◆ ToolType

override eVision.EasyDeepLearning.EDeepLearningToolType eVision.EasyDeepLearning.EClassifier.ToolType

get

Type of the deep learning tool.

◆ UsePretrainedModel

bool eVision.EasyDeepLearning.EClassifier.UsePretrainedModel

getset

Whether to use a pre-trained model when training.

Default: true if the pre-trained model are found on the disk.

◆ Width

uint eVision.EasyDeepLearning.EClassifier.Width

getset

Width for input images of the classifier. By default, this value will be set to 224 if it wasn't explicitly set before training.

If the classifier is not trained or the value was not explicitly set, its value will be 0.

Public Member Functions

Static Public Member Functions

Properties

Detailed Description

Constructor & Destructor Documentation

◆ EClassifier() [1/2]

◆ EClassifier() [2/2]

Member Function Documentation

◆ Classify() [1/12]

◆ Classify() [2/12]

◆ Classify() [3/12]

◆ Classify() [4/12]

◆ Classify() [5/12]

◆ Classify() [6/12]

◆ Classify() [7/12]

◆ Classify() [8/12]

◆ Classify() [9/12]

◆ Classify() [10/12]

◆ Classify() [11/12]

◆ Classify() [12/12]

◆ Evaluate()

◆ GetAvailableModelTypes()

◆ GetColorizedHeatMap()

◆ GetColorizedHeatmapWithTransparency() [1/2]

◆ GetColorizedHeatmapWithTransparency() [2/2]

◆ GetHeatMap()

◆ GetTrainingMetrics()

◆ GetValidationMetrics()

◆ HasOODParameters()

◆ HasPretrainedModel()

◆ LearnOODParameters() [1/5]

◆ LearnOODParameters() [2/5]

◆ LearnOODParameters() [3/5]

◆ LearnOODParameters() [4/5]

◆ LearnOODParameters() [5/5]

◆ LoadAsPretrained()

◆ LoadOnnxModel()

◆ LoadOnnxModelAsPretrained()

◆ SerializeSettings()

◆ Train() [1/8]

◆ Train() [2/8]

◆ Train() [3/8]

◆ Train() [4/8]

◆ Train() [5/8]

◆ Train() [6/8]

◆ Train() [7/8]

◆ Train() [8/8]

Property Documentation

◆ Capacity

◆ Channels

◆ ComputeHeatmapWithResult

◆ ComputeOODWithResult

◆ EnableAutomaticImageReformat

◆ EnableHistogramEqualization

◆ Height

◆ MinimumHeight

◆ MinimumWidth

◆ ModelType

◆ NumAvailableModelTypes

◆ OODThreshold

◆ ToolType

◆ UsePretrainedModel

◆ Width