pulumi-azure-native-kotlin/com.pulumi.azurenative.machinelearningservices.kotlin.outputs/ImageModelSettingsObjectDetectionResponse

ImageModelSettingsObjectDetectionResponse

data class ImageModelSettingsObjectDetectionResponse(val advancedSettings: String? = null, val amsGradient: Boolean? = null, val augmentations: String? = null, val beta1: Double? = null, val beta2: Double? = null, val boxDetectionsPerImage: Int? = null, val boxScoreThreshold: Double? = null, val checkpointFrequency: Int? = null, val checkpointModel: MLFlowModelJobInputResponse? = null, val checkpointRunId: String? = null, val distributed: Boolean? = null, val earlyStopping: Boolean? = null, val earlyStoppingDelay: Int? = null, val earlyStoppingPatience: Int? = null, val enableOnnxNormalization: Boolean? = null, val evaluationFrequency: Int? = null, val gradientAccumulationStep: Int? = null, val imageSize: Int? = null, val layersToFreeze: Int? = null, val learningRate: Double? = null, val learningRateScheduler: String? = null, val maxSize: Int? = null, val minSize: Int? = null, val modelName: String? = null, val modelSize: String? = null, val momentum: Double? = null, val multiScale: Boolean? = null, val nesterov: Boolean? = null, val nmsIouThreshold: Double? = null, val numberOfEpochs: Int? = null, val numberOfWorkers: Int? = null, val optimizer: String? = null, val randomSeed: Int? = null, val stepLRGamma: Double? = null, val stepLRStepSize: Int? = null, val tileGridSize: String? = null, val tileOverlapRatio: Double? = null, val tilePredictionsNmsThreshold: Double? = null, val trainingBatchSize: Int? = null, val validationBatchSize: Int? = null, val validationIouThreshold: Double? = null, val validationMetricType: String? = null, val warmupCosineLRCycles: Double? = null, val warmupCosineLRWarmupEpochs: Int? = null, val weightDecay: Double? = null)

Settings used for training the model. For more information on the available settings please visit the official documentation: https://docs.microsoft.com/en-us/azure/machine-learning/how-to-auto-train-image-models.

Constructors

fun ImageModelSettingsObjectDetectionResponse(advancedSettings: String? = null, amsGradient: Boolean? = null, augmentations: String? = null, beta1: Double? = null, beta2: Double? = null, boxDetectionsPerImage: Int? = null, boxScoreThreshold: Double? = null, checkpointFrequency: Int? = null, checkpointModel: MLFlowModelJobInputResponse? = null, checkpointRunId: String? = null, distributed: Boolean? = null, earlyStopping: Boolean? = null, earlyStoppingDelay: Int? = null, earlyStoppingPatience: Int? = null, enableOnnxNormalization: Boolean? = null, evaluationFrequency: Int? = null, gradientAccumulationStep: Int? = null, imageSize: Int? = null, layersToFreeze: Int? = null, learningRate: Double? = null, learningRateScheduler: String? = null, maxSize: Int? = null, minSize: Int? = null, modelName: String? = null, modelSize: String? = null, momentum: Double? = null, multiScale: Boolean? = null, nesterov: Boolean? = null, nmsIouThreshold: Double? = null, numberOfEpochs: Int? = null, numberOfWorkers: Int? = null, optimizer: String? = null, randomSeed: Int? = null, stepLRGamma: Double? = null, stepLRStepSize: Int? = null, tileGridSize: String? = null, tileOverlapRatio: Double? = null, tilePredictionsNmsThreshold: Double? = null, trainingBatchSize: Int? = null, validationBatchSize: Int? = null, validationIouThreshold: Double? = null, validationMetricType: String? = null, warmupCosineLRCycles: Double? = null, warmupCosineLRWarmupEpochs: Int? = null, weightDecay: Double? = null)

Types

Companion

object Companion

Properties

advancedSettings

val advancedSettings: String? = null

Settings for advanced scenarios.

amsGradient

val amsGradient: Boolean? = null

Enable AMSGrad when optimizer is 'adam' or 'adamw'.

augmentations

val augmentations: String? = null

Settings for using Augmentations.

beta1

val beta1: Double? = null

Value of 'beta1' when optimizer is 'adam' or 'adamw'. Must be a float in the range 0, 1.

beta2

val beta2: Double? = null

Value of 'beta2' when optimizer is 'adam' or 'adamw'. Must be a float in the range 0, 1.

boxDetectionsPerImage

val boxDetectionsPerImage: Int? = null

Maximum number of detections per image, for all classes. Must be a positive integer. Note: This settings is not supported for the 'yolov5' algorithm.

boxScoreThreshold

val boxScoreThreshold: Double? = null

During inference, only return proposals with a classification score greater than BoxScoreThreshold. Must be a float in the range0, 1.

checkpointFrequency

val checkpointFrequency: Int? = null

Frequency to store model checkpoints. Must be a positive integer.

checkpointModel

val checkpointModel: MLFlowModelJobInputResponse? = null

The pretrained checkpoint model for incremental training.

checkpointRunId

val checkpointRunId: String? = null

The id of a previous run that has a pretrained checkpoint for incremental training.

distributed

val distributed: Boolean? = null

Whether to use distributed training.

earlyStopping

val earlyStopping: Boolean? = null

Enable early stopping logic during training.

earlyStoppingDelay

val earlyStoppingDelay: Int? = null

Minimum number of epochs or validation evaluations to wait before primary metric improvement is tracked for early stopping. Must be a positive integer.

earlyStoppingPatience

val earlyStoppingPatience: Int? = null

Minimum number of epochs or validation evaluations with no primary metric improvement before the run is stopped. Must be a positive integer.

enableOnnxNormalization

val enableOnnxNormalization: Boolean? = null

Enable normalization when exporting ONNX model.

evaluationFrequency

val evaluationFrequency: Int? = null

Frequency to evaluate validation dataset to get metric scores. Must be a positive integer.

gradientAccumulationStep

val gradientAccumulationStep: Int? = null

Gradient accumulation means running a configured number of "GradAccumulationStep" steps without updating the model weights while accumulating the gradients of those steps, and then using the accumulated gradients to compute the weight updates. Must be a positive integer.

imageSize

val imageSize: Int? = null

Image size for train and validation. Must be a positive integer. Note: The training run may get into CUDA OOM if the size is too big. Note: This settings is only supported for the 'yolov5' algorithm.

layersToFreeze

val layersToFreeze: Int? = null

Number of layers to freeze for the model. Must be a positive integer. For instance, passing 2 as value for 'seresnext' means freezing layer0 and layer1. For a full list of models supported and details on layer freeze, please see: https://docs.microsoft.com/en-us/azure/machine-learning/how-to-auto-train-image-models.

learningRate

val learningRate: Double? = null

Initial learning rate. Must be a float in the range 0, 1.

learningRateScheduler

val learningRateScheduler: String? = null

Type of learning rate scheduler. Must be 'warmup_cosine' or 'step'.

maxSize

val maxSize: Int? = null

Maximum size of the image to be rescaled before feeding it to the backbone. Must be a positive integer. Note: training run may get into CUDA OOM if the size is too big. Note: This settings is not supported for the 'yolov5' algorithm.

minSize

val minSize: Int? = null

Minimum size of the image to be rescaled before feeding it to the backbone. Must be a positive integer. Note: training run may get into CUDA OOM if the size is too big. Note: This settings is not supported for the 'yolov5' algorithm.

modelName

val modelName: String? = null

Name of the model to use for training. For more information on the available models please visit the official documentation: https://docs.microsoft.com/en-us/azure/machine-learning/how-to-auto-train-image-models.

modelSize

val modelSize: String? = null

Model size. Must be 'small', 'medium', 'large', or 'xlarge'. Note: training run may get into CUDA OOM if the model size is too big. Note: This settings is only supported for the 'yolov5' algorithm.

momentum

val momentum: Double? = null

Value of momentum when optimizer is 'sgd'. Must be a float in the range 0, 1.

multiScale

val multiScale: Boolean? = null

Enable multi-scale image by varying image size by +/- 50%. Note: training run may get into CUDA OOM if no sufficient GPU memory. Note: This settings is only supported for the 'yolov5' algorithm.

nesterov

val nesterov: Boolean? = null

Enable nesterov when optimizer is 'sgd'.

nmsIouThreshold

val nmsIouThreshold: Double? = null

IOU threshold used during inference in NMS post processing. Must be a float in the range 0, 1.

numberOfEpochs

val numberOfEpochs: Int? = null

Number of training epochs. Must be a positive integer.

numberOfWorkers

val numberOfWorkers: Int? = null

Number of data loader workers. Must be a non-negative integer.

optimizer

val optimizer: String? = null

Type of optimizer.

randomSeed

val randomSeed: Int? = null

Random seed to be used when using deterministic training.

stepLRGamma

val stepLRGamma: Double? = null

Value of gamma when learning rate scheduler is 'step'. Must be a float in the range 0, 1.

stepLRStepSize

val stepLRStepSize: Int? = null

Value of step size when learning rate scheduler is 'step'. Must be a positive integer.

tileGridSize

val tileGridSize: String? = null

The grid size to use for tiling each image. Note: TileGridSize must not be None to enable small object detection logic. A string containing two integers in mxn format. Note: This settings is not supported for the 'yolov5' algorithm.

tileOverlapRatio

val tileOverlapRatio: Double? = null

Overlap ratio between adjacent tiles in each dimension. Must be float in the range [0, 1). Note: This settings is not supported for the 'yolov5' algorithm.

tilePredictionsNmsThreshold

val tilePredictionsNmsThreshold: Double? = null

The IOU threshold to use to perform NMS while merging predictions from tiles and image. Used in validation/ inference. Must be float in the range 0, 1. Note: This settings is not supported for the 'yolov5' algorithm.

trainingBatchSize

val trainingBatchSize: Int? = null

Training batch size. Must be a positive integer.

validationBatchSize

val validationBatchSize: Int? = null

Validation batch size. Must be a positive integer.

validationIouThreshold

val validationIouThreshold: Double? = null

IOU threshold to use when computing validation metric. Must be float in the range 0, 1.

validationMetricType

val validationMetricType: String? = null

Metric computation method to use for validation metrics.

warmupCosineLRCycles

val warmupCosineLRCycles: Double? = null

Value of cosine cycle when learning rate scheduler is 'warmup_cosine'. Must be a float in the range 0, 1.

warmupCosineLRWarmupEpochs

val warmupCosineLRWarmupEpochs: Int? = null

Value of warmup epochs when learning rate scheduler is 'warmup_cosine'. Must be a positive integer.

weightDecay

val weightDecay: Double? = null

Value of weight decay when optimizer is 'sgd', 'adam', or 'adamw'. Must be a float in the range0, 1.