Normal(mu, sigma) - Returns a real value that's normally distributed with mean mu and standard deviation sigma.

LogUniform(min_value, max_value) - Returns a value drawn according to exp(Uniform(min_value, max_value)) so that the logarithm of the return value is uniformly distributed.

Uniform(min_value, max_value) - Returns a value uniformly distributed between min_value and max_value.

The Continuous hyperparameters are specified as a distribution over a continuous range of values:

QLogNormal(mu, sigma, q) - Returns a value like round(exp(Normal(mu, sigma)) / q) * q.QNormal(mu, sigma, q) - Returns a value like round(Normal(mu, sigma) / q) * q.QLogUniform(min_value, max_value, q) - Returns a value like round(exp(Uniform(min_value, max_value)) / q) * q.QUniform(min_value, max_value, q) - Returns a value like round(Uniform(min_value, max_value) / q) * q.The following advanced discrete hyperparameters can also be specified using a distribution: In this case, batch_size one of the values and number_of_hidden_layers takes one of the values. Number_of_hidden_layers=Choice(values=range(1,5)), Choice can be:īatch_size=Choice(values=), Discrete hyperparametersĭiscrete hyperparameters are specified as a Choice among discrete values. Hyperparameters can be discrete or continuous, and has a distribution of values described by a Tune hyperparameters by exploring the range of values defined for each hyperparameter. The process is typically computationally expensive and manual.Īzure Machine Learning lets you automate hyperparameter tuning and run experiments in parallel to efficiently optimize hyperparameters. Hyperparameter tuning, also called hyperparameter optimization, is the process of finding the configuration of hyperparameters that results in the best performance. Model performance depends heavily on hyperparameters. For example, with neural networks, you decide the number of hidden layers and the number of nodes in each layer. Hyperparameters are adjustable parameters that let you control the model training process. Select the best configuration for your model.Launch an experiment with the defined configuration.Specify early termination policy for low-performing jobs.

Specify the sampling algorithm for your sweep job.Define the parameter search space for your trial.Automate efficient hyperparameter tuning using Azure Machine Learning SDK v2 and CLI v2 by way of the SweepJob type.

Hyperdrive net