Exam DP-100 topic 3 question 86 discussion

The correct answer is B and D.

The idea here is that, ABC are hyperparameters that Azure can figure out for you, but it needs D and E. why D? Because Azure needs to know when to stop. I.e. it can't run forever why E? Because you need to tell Azure what to "sweep over". Is it to sweep over hidden layer breadth? Depth? both? for each of these sweep runs, it tries a permutation of learning rate and kernel function etc

The correct answer is B and D.

BDE https://learn.microsoft.com/en-us/azure/machine-learning/component-reference/two-class-neural-network?view=azureml-api-2

A. Number of hidden nodes B. Learning Rate

A. Number of hidden nodes B. Learning Rate

B and D

A. Number of hidden nodes B. Learning Rate When configuring the Tune Model Hyperparameters module to tune accuracy for a Two-Class Neural Network model in Azure Machine Learning Studio, you should select the hyperparameters you want to tune. In this case, you should select the number of hidden nodes and the learning rate. These are two of the most important hyperparameters for a neural network and can significantly impact the accuracy of the model. The other options listed (C, D, and E) are not relevant hyperparameters for the Two-Class Neural Network module. The type of normalizer is a preprocessing step and is not a hyperparameter for the model. The number of learning iterations and the hidden layer specification are both set by the Two-Class Neural Network module and cannot be directly tuned using the Tune Model Hyperparameters module.

Learning rate and hidden layer specification. No idea why (ironically) the community vote is tied at 50-50 on question about binary classification :)

As per Parimt's explanation

Answer only D? Other options are not in properties of Tune Model Hyperparameters . Tune Model Hyperparameters Properties: Sweep Method Maximum number of runs Random seed Label Column Metric https://learn.microsoft.com/zh-tw/previous-versions/azure/machine-learning/studio-module-reference/tune-model-hyperparameters

Yes correct ans looks to be B and D as when I ran a small expt in designer using tune hyperparameter module and nn, I get the below output for sweep data preview with these column headers- Number of learning iterations Learning rate mean_test_Mean absolute error mean_test_Root of mean squared error mean_test_Relative absolute error mean_test_Relative squared error mean_test_Coefficient of determination rank

B and D

the answer is 'B' and 'D' if 'D' and 'E' are correct then 'A' i also has to be correct.

B and D: The 2-class Neural Network has only 2 parameters that can be set to "range" (instead of "single parameter"), which in turn can be learned by the Tune Model Hyperparameters module: 1) number of iterations 2) Learning rate.

B and D Below some text from the "Two-Class Neural Network module": 4. For "Learning rate", define the size of the step taken at each iteration, before correction. A larger value for learning rate can cause the model to converge faster, but it can overshoot local minima. 5. For "Number of learning iterations", specify the maximum number of times the algorithm should process the training cases.

I think is B and D