ArmMachineLearningModelFactory.ImageModelDistributionSettingsClassification 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.ImageModelDistributionSettingsClassification ImageModelDistributionSettingsClassification(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 trainingCropSize = default, string validationCropSize = default, string validationResizeSize = default, string weightedLoss = default);

static member ImageModelDistributionSettingsClassification : 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.ImageModelDistributionSettingsClassification

Public Shared Function ImageModelDistributionSettingsClassification (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 trainingCropSize As String = Nothing, Optional validationCropSize As String = Nothing, Optional validationResizeSize As String = Nothing, Optional weightedLoss As String = Nothing) As ImageModelDistributionSettingsClassification

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].

trainingCropSize: String

Image crop size that is input to the neural network for the training dataset. Must be a positive integer.

validationCropSize: String

Image crop size that is input to the neural network for the validation dataset. Must be a positive integer.

validationResizeSize: String

Image size to which to resize before cropping for validation dataset. Must be a positive integer.

weightedLoss: String

Weighted loss. The accepted values are 0 for no weighted loss. 1 for weighted loss with sqrt.(class_weights). 2 for weighted loss with class_weights. Must be 0 or 1 or 2.

Returns

ImageModelDistributionSettingsClassification

A new ImageModelDistributionSettingsClassification instance for mocking.

Applies to

Commenti e suggerimenti

Questa pagina è stata utile?