Deep Learning

HALCON ScriptCpp++.NETPythonC

On this topic

• Deep Learning

• Deep Learning – Anomaly Detection

• Deep Learning – Classification

• Deep Learning – Continual Learning

• Deep Learning – Instance Segmentation

• Deep Learning – Model

• Deep Learning – Multi Label Classification

• Deep Learning – Object Detection

• Deep Learning – Semantic Segmentation

General

create_dl_transform_blurCreateDlTransformBlur: Create a blur transform.

create_dl_transform_flipCreateDlTransformFlip: Create a flip transform.

create_dl_transform_normalizeCreateDlTransformNormalize: Create a normalize transform.

create_dl_transform_pipelineCreateDlTransformPipeline: Create a pipeline of transforms for the training of a deep learning model.

create_dl_transform_random_cropCreateDlTransformRandomCrop: Create a random crop transform.

create_dl_transform_random_geometricCreateDlTransformRandomGeometric: Create a random geometric transform.

create_dl_transform_random_hsvCreateDlTransformRandomHsv: Create a random HSV transform.

create_dl_transform_resizeCreateDlTransformResize: Create a resize transform using absolute dimensions.

get_dl_device_paramGetDlDeviceParam: Return the parameters of a deep-learning-capable hardware device.

get_dl_transform_pipeline_paramGetDlTransformPipelineParam: Retrieve parameter values for a given transform in the transform pipeline, or for the transform pipeline itself.

optimize_dl_model_for_inferenceOptimizeDlModelForInference: Optimize a model for inference on a device via the AI²-interface.

query_available_dl_devicesQueryAvailableDlDevices: Get list of deep-learning-capable hardware devices.

read_dl_transform_pipelineReadDlTransformPipeline: Read a deep learning transform pipeline from a file.

set_dl_device_paramSetDlDeviceParam: Set the parameters of a deep-learning-capable hardware device.

set_dl_transform_pipeline_paramSetDlTransformPipelineParam: Set parameter values for a given transform in the transform pipeline, or for the transform pipeline itself.

transform_dl_sample_batchTransformDlSampleBatch: Apply a pipeline of transforms on a set of images.

write_dl_transform_pipelineWriteDlTransformPipeline: Write a deep learning transform pipeline to a file

Anomaly Detection

train_dl_model_anomaly_datasetTrainDlModelAnomalyDataset: Train a deep learning model for anomaly detection.

Classification

fit_dl_out_of_distributionFitDlOutOfDistribution: Extend a deep learning model for Out-of-Distribution Detection.

Continual Learning

extend_dl_continual_learningExtendDlContinualLearning: Extend a Continual Learning model with with new data.

init_dl_continual_learningInitDlContinualLearning: Converts a deep learning model to a model suitable for Continual Learning.

Framework

create_dl_layer_activationCreateDlLayerActivation: Create an activation layer.

create_dl_layer_affine_gridCreateDlLayerAffineGrid: Create an affine grid layer.

create_dl_layer_anchorsCreateDlLayerAnchors: Create a layer for generating anchor boxes.

create_dl_layer_batch_normalizationCreateDlLayerBatchNormalization: Create a batch normalization layer.

create_dl_layer_box_proposalsCreateDlLayerBoxProposals: Create a layer for generating box proposals.

create_dl_layer_box_targetsCreateDlLayerBoxTargets: Create a layer for generating box targets.

create_dl_layer_class_id_conversionCreateDlLayerClassIdConversion: Create a class ID conversion layer.

create_dl_layer_concatCreateDlLayerConcat: Create a concatenation layer.

create_dl_layer_convolutionCreateDlLayerConvolution: Create a convolutional layer.

create_dl_layer_denseCreateDlLayerDense: Create a dense layer.

create_dl_layer_depth_maxCreateDlLayerDepthMax: Create a depth max layer.

create_dl_layer_depth_to_spaceCreateDlLayerDepthToSpace: Create a depth to space layer.

create_dl_layer_dropoutCreateDlLayerDropout: Create a DropOut layer.

create_dl_layer_elementwiseCreateDlLayerElementwise: Create an elementwise layer.

create_dl_layer_gatherCreateDlLayerGather: Create a gather layer.

create_dl_layer_grid_sampleCreateDlLayerGridSample: Create a grid sample layer.

create_dl_layer_identityCreateDlLayerIdentity: Create an identity layer.

create_dl_layer_inputCreateDlLayerInput: Create an input layer.

create_dl_layer_loss_cross_entropyCreateDlLayerLossCrossEntropy: Create a cross entropy loss layer.

create_dl_layer_loss_ctcCreateDlLayerLossCtc: Create a CTC loss layer.

create_dl_layer_loss_distanceCreateDlLayerLossDistance: Create a distance loss layer.

create_dl_layer_loss_focalCreateDlLayerLossFocal: Create a focal loss layer.

create_dl_layer_loss_huberCreateDlLayerLossHuber: Create a Huber loss layer.

create_dl_layer_lrnCreateDlLayerLrn: Create a LRN layer.

create_dl_layer_matmulCreateDlLayerMatmul: Create a MatMul layer.

create_dl_layer_permutationCreateDlLayerPermutation: Create a permutation layer.

create_dl_layer_poolingCreateDlLayerPooling: Create a pooling layer.

create_dl_layer_reduceCreateDlLayerReduce: Create a reduce layer.

create_dl_layer_reshapeCreateDlLayerReshape: Create a reshape layer.

create_dl_layer_roi_poolingCreateDlLayerRoiPooling: Create an ROI pooling layer.

create_dl_layer_sliceCreateDlLayerSlice: Create a slice layer.

create_dl_layer_softmaxCreateDlLayerSoftmax: Create a softmax layer.

create_dl_layer_transposed_convolutionCreateDlLayerTransposedConvolution: Create a transposed convolution layer.

create_dl_layer_zoom_factorCreateDlLayerZoomFactor: Create a zoom layer using size factors.

create_dl_layer_zoom_sizeCreateDlLayerZoomSize: Create a zoom layer using an absolute output size.

create_dl_layer_zoom_to_layer_sizeCreateDlLayerZoomToLayerSize: Create a zoom layer using the output size of a reference layer.

create_dl_modelCreateDlModel: Create a deep learning model.

get_dl_layer_paramGetDlLayerParam: Return the parameters of a deep learning layer.

get_dl_model_layerGetDlModelLayer: Create a deep copy of the layers and all of their graph ancestors in a given deep learning model.

get_dl_model_layer_activationsGetDlModelLayerActivations: Get the activations of a Deep Learning model layer.

get_dl_model_layer_gradientsGetDlModelLayerGradients: Get the gradients of a Deep Learning model layer.

get_dl_model_layer_paramGetDlModelLayerParam: Retrieve parameter values for a given layer.

get_dl_model_layer_weightsGetDlModelLayerWeights: Get the weights (or values) of a Deep Learning model layer.

load_dl_model_weightsLoadDlModelWeights: Load the weights of a source model into a target model.

set_dl_model_layer_paramSetDlModelLayerParam: Set parameter values of a given layer.

set_dl_model_layer_weightsSetDlModelLayerWeights: Set the weights (or values) of a Deep Learning model layer.

Instance Segmentation

create_dl_model_detectionCreateDlModelDetection: Create a deep learning network for object detection or instance segmentation.

Model

add_dl_pruning_batchAddDlPruningBatch: Calculate scores to prune a deep learning model.

apply_dl_modelApplyDlModel: Apply a deep-learning-based network on a set of images for inference.

clear_dl_modelClearDlModel: Clear a deep learning model.

create_dl_pruningCreateDlPruning: Create a pruning data handle.

deserialize_dl_modelDeserializeDlModel: Deserialize a deep learning model.

gen_dl_model_heatmapGenDlModelHeatmap: Infer the sample and generate a heatmap.

gen_dl_pruned_modelGenDlPrunedModel: Prune a deep learning model.

get_dl_model_paramGetDlModelParam: Return the parameters of a deep learning model.

get_dl_pruning_paramGetDlPruningParam: Get information from a pruning data handle.

read_dl_modelReadDlModel: Read a deep learning model from a file.

serialize_dl_modelSerializeDlModel: Serialize a deep learning model.

set_dl_model_paramSetDlModelParam: Set the parameters of a deep learning model.

set_dl_pruning_paramSetDlPruningParam: Set parameter in a pruning data handle.

train_dl_model_batchTrainDlModelBatch: Train a deep learning model.

write_dl_modelWriteDlModel: Write a deep learning model in a file.