Databricks Exam Databricks Machine Learning Associate Topic 2 Question 19 Discussion

Actual exam question for Databricks's Databricks Machine Learning Associate exam

Question #: 19
Topic #: 2

[All Databricks Machine Learning Associate Questions]

A data scientist wants to tune a set of hyperparameters for a machine learning model. They have wrapped a Spark ML model in the objective function objective_function and they have defined the search space search_space.

As a result, they have the following code block:

Which of the following changes do they need to make to the above code block in order to accomplish the task?

AChange SparkTrials() to Trials()

BReduce num_evals to be less than 10

CChange fmin() to fmax()

DRemove the trials=trials argument

ERemove the algo=tpe.suggest argument

Show Suggested Answer

Suggested Answer: C

To find the run_id of the run with the best root-mean-square error (RMSE) in an MLflow experiment, the correct line of code to use is:

mlflow.search_runs( experiment_id, order_by=['metrics.rmse'] )['run_id'][0]

This line of code searches the runs in the specified experiment, orders them by the RMSE metric in ascending order (the lower the RMSE, the better), and retrieves the run_id of the best-performing run. Option C correctly represents this logic.

Reference

MLflow documentation on tracking experiments: https://www.mlflow.org/docs/latest/python_api/mlflow.html#mlflow.search_runs

by Rebbecca at Sep 15, 2024, 08:23 PM

Limited Time Offer

25%

Off

Get Premium Databricks Machine Learning Associate Questions as Interactive Web-Based Practice Test or PDF

Contribute your Thoughts:

Submit Cancel

Lenita

1 months ago

Wait, they're tuning hyperparameters for a machine learning model? I thought this was a cooking show. Where's the recipe for the perfect hyperparameter pie?

upvoted 0 times

...

Leonie

1 months ago

Hmm, I'm not sure about option D) or E). Removing the trials=trials argument or the algo=tpe.suggest argument seems like it would break the code. I think I'll go with option A).

upvoted 0 times

Micah

1 days ago

I agree, changing SparkTrials() to Trials() makes sense.

upvoted 0 times

...

Dierdre

14 days ago

I think option A) is the best choice.

upvoted 0 times

...

Celeste

2 months ago

Hold up, did they really ask us to change fmin() to fmax()? That's like asking a chef to bake a cake by setting the oven to broil. Option C) is definitely not the correct answer.

upvoted 0 times

Rosendo

23 days ago

D) Remove the trials=trials argument

upvoted 0 times

...

Tayna

27 days ago

B) Reduce num_evals to be less than 10

upvoted 0 times

...

Vernell

1 months ago

A) Change SparkTrials() to Trials()

upvoted 0 times

...

Nickolas

2 months ago

I think option B) is the way to go. Reducing the number of evaluations can save time and resources during the hyperparameter tuning process.

upvoted 0 times

Colette

20 days ago

I think focusing on reducing the number of evaluations is more important than changing the algorithm suggestion.

upvoted 0 times

...

Natalie

22 days ago

But what if we also remove the algo=tpe.suggest argument? Wouldn't that help as well?

upvoted 0 times

...

Allene

2 months ago

I agree, reducing the number of evaluations can speed up the tuning process.

upvoted 0 times

...

Lauran

2 months ago

The correct answer is A) Change SparkTrials() to Trials(). The question specifies that the data scientist is using a Spark ML model, so they need to use the SparkTrials() function instead of the regular Trials() function.

upvoted 0 times