ArmMachineLearningModelFactory.ImageModelDistributionSettingsObjectDetection Method

Definition

Namespace:: Azure.ResourceManager.MachineLearning.Models

Assembly:: Azure.ResourceManager.MachineLearning.dll

Package:: Azure.ResourceManager.MachineLearning v1.3.0

Source:: ArmMachineLearningModelFactory.cs

Important

Some information relates to prerelease product that may be substantially modified before it’s released. Microsoft makes no warranties, express or implied, with respect to the information provided here.

Distribution expressions to sweep over values of model settings. <example> Some examples are:

ModelName = "choice('seresnext', 'resnest50')";
LearningRate = "uniform(0.001, 0.01)";
LayersToFreeze = "choice(0, 2)";
```&lt;/example&gt;
For more details on how to compose distribution expressions please check the documentation:
https://docs.microsoft.com/en-us/azure/machine-learning/how-to-tune-hyperparameters
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.

public static Azure.ResourceManager.MachineLearning.Models.ImageModelDistributionSettingsObjectDetection ImageModelDistributionSettingsObjectDetection(string amsGradient = default, string augmentations = default, string beta1 = default, string beta2 = default, string distributed = default, string earlyStopping = default, string earlyStoppingDelay = default, string earlyStoppingPatience = default, string enableOnnxNormalization = default, string evaluationFrequency = default, string gradientAccumulationStep = default, string layersToFreeze = default, string learningRate = default, string learningRateScheduler = default, string modelName = default, string momentum = default, string nesterov = default, string numberOfEpochs = default, string numberOfWorkers = default, string optimizer = default, string randomSeed = default, string stepLRGamma = default, string stepLRStepSize = default, string trainingBatchSize = default, string validationBatchSize = default, string warmupCosineLRCycles = default, string warmupCosineLRWarmupEpochs = default, string weightDecay = default, string boxDetectionsPerImage = default, string boxScoreThreshold = default, string imageSize = default, string maxSize = default, string minSize = default, string modelSize = default, string multiScale = default, string nmsIouThreshold = default, string tileGridSize = default, string tileOverlapRatio = default, string tilePredictionsNmsThreshold = default, string validationIouThreshold = default, string validationMetricType = default);

static member ImageModelDistributionSettingsObjectDetection : string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string * string -> Azure.ResourceManager.MachineLearning.Models.ImageModelDistributionSettingsObjectDetection

Public Shared Function ImageModelDistributionSettingsObjectDetection (Optional amsGradient As String = Nothing, Optional augmentations As String = Nothing, Optional beta1 As String = Nothing, Optional beta2 As String = Nothing, Optional distributed As String = Nothing, Optional earlyStopping As String = Nothing, Optional earlyStoppingDelay As String = Nothing, Optional earlyStoppingPatience As String = Nothing, Optional enableOnnxNormalization As String = Nothing, Optional evaluationFrequency As String = Nothing, Optional gradientAccumulationStep As String = Nothing, Optional layersToFreeze As String = Nothing, Optional learningRate As String = Nothing, Optional learningRateScheduler As String = Nothing, Optional modelName As String = Nothing, Optional momentum As String = Nothing, Optional nesterov As String = Nothing, Optional numberOfEpochs As String = Nothing, Optional numberOfWorkers As String = Nothing, Optional optimizer As String = Nothing, Optional randomSeed As String = Nothing, Optional stepLRGamma As String = Nothing, Optional stepLRStepSize As String = Nothing, Optional trainingBatchSize As String = Nothing, Optional validationBatchSize As String = Nothing, Optional warmupCosineLRCycles As String = Nothing, Optional warmupCosineLRWarmupEpochs As String = Nothing, Optional weightDecay As String = Nothing, Optional boxDetectionsPerImage As String = Nothing, Optional boxScoreThreshold As String = Nothing, Optional imageSize As String = Nothing, Optional maxSize As String = Nothing, Optional minSize As String = Nothing, Optional modelSize As String = Nothing, Optional multiScale As String = Nothing, Optional nmsIouThreshold As String = Nothing, Optional tileGridSize As String = Nothing, Optional tileOverlapRatio As String = Nothing, Optional tilePredictionsNmsThreshold As String = Nothing, Optional validationIouThreshold As String = Nothing, Optional validationMetricType As String = Nothing) As ImageModelDistributionSettingsObjectDetection

Parameters

amsGradient: String

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

augmentations: String

Settings for using Augmentations.

beta1: String

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

beta2: String

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

distributed: String

Whether to use distributer training.

earlyStopping: String

Enable early stopping logic during training.

earlyStoppingDelay: String

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

earlyStoppingPatience: String

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

enableOnnxNormalization: String

Enable normalization when exporting ONNX model.

evaluationFrequency: String

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

gradientAccumulationStep: String

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.

layersToFreeze: String

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: String

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

learningRateScheduler: String

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

modelName: String

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.

momentum: String

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

nesterov: String

Enable nesterov when optimizer is 'sgd'.

numberOfEpochs: String

Number of training epochs. Must be a positive integer.

numberOfWorkers: String

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

optimizer: String

Type of optimizer. Must be either 'sgd', 'adam', or 'adamw'.

randomSeed: String

Random seed to be used when using deterministic training.

stepLRGamma: String

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

stepLRStepSize: String

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

trainingBatchSize: String

Training batch size. Must be a positive integer.

validationBatchSize: String

Validation batch size. Must be a positive integer.

warmupCosineLRCycles: String

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

warmupCosineLRWarmupEpochs: String

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

weightDecay: String

Value of weight decay when optimizer is 'sgd', 'adam', or 'adamw'. Must be a float in the range[0, 1].

boxDetectionsPerImage: String

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: String

During inference, only return proposals with a classification score greater than BoxScoreThreshold. Must be a float in the range[0, 1].

imageSize: String

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.

maxSize: String

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: String

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.

modelSize: String

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.

multiScale: String

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.

nmsIouThreshold: String

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

tileGridSize: String

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: String

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: String

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. NMS: Non-maximum suppression

validationIouThreshold: String

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

validationMetricType: String

Metric computation method to use for validation metrics. Must be 'none', 'coco', 'voc', or 'coco_voc'.

Returns

ImageModelDistributionSettingsObjectDetection

A new ImageModelDistributionSettingsObjectDetection instance for mocking.

Applies to

Feedback

Var denne side nyttig?