Exam DP-100 topic 3 question 12 discussion

Box 2 should be NO as 'Policy' parameter is for early termination policy not security policy.

So Answer could be 3 NO s

Please verify the information in the solution section to this question. The actual answer is 1. Yes. The max_total_runs parameter is set to 50, so the hyperparameter tuning experiment will have up to 50 total iterations. 2. No. The policy parameter in HyperDriveConfig relates to the early termination policy, not security policies. It defines stopping rules for low-performing runs. 3. No. The learning rate parameter uses GridParameterSampling with discrete values specified (0.05, 0.1, 0.15). So only 3 options listed would be searched, not the full range.

Can someone explain why thrid box is a no when grid sampling tries every possible option?

First box Yes. Second Box No, Third Box No. For the first box, the total number of combinations of max_depth and learning rate is 81. However runs will be capped to 50, by max_runs parameter.

Its No, No and No

No, No, No, Policy = Early Termination Policy

similar question on 18-5-22

Box 2 should be NO

On march 2022

Box1 No Box2 No Box3 No

very funny answer. box 1 is yes, 2 is no, 3 is no

box 2 is No https://docs.microsoft.com/en-us/python/api/azureml-train-core/azureml.train.hyperdrive.hyperdriveconfig?view=azure-ml-py Policy: The early termination policy to use. If None - the default, no early termination policy will be used.