Free Databricks Certified Machine Learning Associate Exam Databricks-Machine-Learning-Associate Exam Practice Test

Which of the following is a benefit of using vectorized pandas UDFs instead of standard PySpark UDFs?

• The vectorized pandas UDFs allow for the use of type hints
• The vectorized pandas UDFs process data in batches rather than one row at a time
• The vectorized pandas UDFs allow for pandas API use inside of the function
• The vectorized pandas UDFs work on distributed DataFrames
• The vectorized pandas UDFs process data in memory rather than spilling to disk

A data scientist has produced three new models for a single machine learning problem. In the past, the solution used just one model. All four models have nearly the same prediction latency, but a machine learning engineer suggests that the new solution will be less time efficient during inference. In which situation will the machine learning engineer be correct?

• When the new solution requires if-else logic determining which model to use to compute each prediction
• When the new solution's models have an average latency that is larger than the size of the original model
• When the new solution requires the use of fewer feature variables than the original model
• When the new solution requires that each model computes a prediction for every record
• When the new solution's models have an average size that is larger than the size of the original model

A data scientist learned during their training to always use 5-fold cross-validation in their model development workflow. A colleague suggests that there are cases where a train-validation split could be preferred over k-fold cross-validation when k > 2.Which of the following describes a potential benefit of using a train-validation split over k-fold cross- validation in this scenario?

• A holdout set is not necessary when using a train-validation split
• Reproducibility is achievable when using a train-validation split
• Fewer hyperparameter values need to be tested when using a train-validation split
• Bias is avoidable when using a train-validation split
• Fewer models need to be trained when using a train-validation split

Which of the following machine learning algorithms typically uses bagging?

• Gradient boosted trees
• K-means
• Random forest
• Linear regression
• Decision tree

A machine learning engineering team has a Job with three successive tasks. Each task runs a single notebook. The team has been alerted that the Job has failed in its latest run.Which of the following approaches can the team use to identify which task is the cause of the failure?

• Run each notebook interactively
• Review the matrix view in the Job's runs
• Migrate the Job to a Delta Live Tables pipeline
• Change each Task’s setting to use a dedicated cluster

Which statement describes a Spark ML transformer?

• A transformer is an algorithm which can transform one DataFrame into another DataFrame
• A transformer is a hyperparameter grid that can be used to train a model
• A transformer chains multiple algorithms together to transform an ML workflow
• A transformer is a learning algorithm that can use a DataFrame to train a model

A data scientist has created two linear regression models. The first model uses price as a label variable and the second model uses log(price) as a label variable. When evaluating the RMSE of each model by comparing the label predictions to the actual price values, the data scientist notices that the RMSE for the second model is much larger than the RMSE of the first model.Which of the following possible explanations for this difference is invalid?

• The second model is much more accurate than the first model
• The data scientist failed to exponentiate the predictions in the second model prior to computing the RMSE
• The data scientist failed to take the log of the predictions in the first model prior to computing the RMSE
• The first model is much more accurate than the second model
• The RMSE is an invalid evaluation metric for regression problems

Which of the following tools can be used to distribute large-scale feature engineering without the use of a UDF or pandas Function API for machine learning pipelines?

• Keras
• pandas
• PvTorch
• Spark ML
• Scikit-learn

A data scientist uses 3-fold cross-validation and the following hyperparameter grid when optimizing model hyperparameters via grid search for a classification problem:Hyperparameter 1: [2, 5, 10]Hyperparameter 2: [50, 100]Which of the following represents the number of machine learning models that can be trained in parallel during this process?

• 3
• 5
• 6
• 18

A data scientist has developed a machine learning pipeline with a static input data set using Spark ML, but the pipeline is taking too long to process. They increase the number of workers in the cluster to get the pipeline to run more efficiently. They notice that the number of rows in the training set after reconfiguring the cluster is different from the number of rows in the training set prior to reconfiguring the cluster.Which of the following approaches will guarantee a reproducible training and test set for each model?

• Manually configure the cluster
• Write out the split data sets to persistent storage
• Set a speed in the data splitting operation
• Manually partition the input data