You use Azure Machine Learning to train a model. You must use Bayesian sampling to tune hyperparameters. You need to select a learning_rate parameter distribution. Which two distributions can you use?

The correct answer is A. Uniform B. Choice. Bayesian sampling in Azure Machine Learning has restrictions on which parameter distributions it supports. It supports uniform (a continuous uniform distribution over a range) and choice (a discrete set of values) because these distributions are compatible with the Gaussian proce

Explore data and run experiments

Question

You use Azure Machine Learning to train a model. You must use Bayesian sampling to tune hyperparameters. You need to select a learning_rate parameter distribution. Which two distributions can you use? Each correct answer presents a complete solution. NOTE: Each correct selection is worth one point.

Options

AUniform
BChoice
CQNormal
DNormal
ELogUniform

How the community answered

(54 responses)

A
89% (48)
C
4% (2)
D
6% (3)
E
2% (1)

Explanation

Bayesian sampling in Azure Machine Learning has restrictions on which parameter distributions it supports. It supports uniform (a continuous uniform distribution over a range) and choice (a discrete set of values) because these distributions are compatible with the Gaussian process surrogate model that Bayesian optimization relies on. QNormal (option C) is a quantized normal distribution that is not supported by Bayesian sampling. Normal (option D) is a Gaussian distribution that is also not supported by Bayesian sampling. LogUniform (option E) is similarly not supported. When using Bayesian sampling, hyperparameters must be defined using only the uniform or choice distributions.

Topics

#Azure Machine Learning#Hyperparameter Tuning#Bayesian Sampling#Parameter Distributions

Community Discussion

No community discussion yet for this question.

Full DP-100 Practice