Databricks Databricks-Machine-Learning-Associate Dumps

Question # 1

Which of the following machine learning algorithms typically uses bagging?

A. IGradient boosted trees
B. K-means
C. Random forest
D. Decision tree

Show Answer

Question # 2

The implementation of linear regression in Spark ML first attempts to solve the linear regressionproblem using matrix decomposition, but this method does not scale well to large datasets with alarge number of variables.Which of the following approaches does Spark ML use to distribute the training of a linear regressionmodel for large data?

A. Logistic regression
B. Singular value decomposition
C. Iterative optimization

Show Answer

Question # 3

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 amachine learning engineer suggests that the new solution will be less time efficient during inference.In which situation will the machine learning engineer be correct?

A. When the new solution requires if-else logic determining which model to use to compute eachprediction
B. When the new solution's models have an average latency that is larger than the size of theoriginal model
C. When the new solution requires the use of fewer feature variables than the original model
D. When the new solution requires that each model computes a prediction for every record
E. When the new solution's models have an average size that is larger than the size of the originalmodel

Show Answer

Question # 4

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

A. Manually configure the cluster
B. Write out the split data sets to persistent storage
C. Set a speed in the data splitting operation
D. Manually partition the input data

Show Answer

Question # 5

A data scientist is developing a single-node machine learning model. They have a large number ofmodel configurations to test as a part of their experiment. As a result, the model tuning processtakes too long to complete. Which of the following approaches can be used to speed up the modeltuning process?

A. Implement MLflow Experiment Tracking
B. Scale up with Spark ML
C. Enable autoscaling clusters
D. Parallelize with Hyperopt

Show Answer

Question # 6

A machine learning engineer is trying to scale a machine learning pipeline by distributing its singlenodemodel tuning process. After broadcasting the entire training data onto each core, each core inthe cluster can train one model at a time. Because the tuning process is still running slowly, theengineer wants to increase the level of parallelism from 4 cores to 8 cores to speed up the tuningprocess. Unfortunately, the total memory in the cluster cannot be increased.In which of the following scenarios will increasing the level of parallelism from 4 to 8 speed up thetuning process?

A. When the tuning process in randomized
B. When the entire data can fit on each core
C. When the model is unable to be parallelized
D. When the data is particularly long in shape
E. When the data is particularly wide in shape

Show Answer

Question # 7

A data scientist has been given an incomplete notebook from the data engineering team. Thenotebook uses a Spark DataFrame spark_df on which the data scientist needs to perform furtherfeature engineering. Unfortunately, the data scientist has not yet learned the PySpark DataFrameAPI.Which of the following blocks of code can the data scientist run to be able to use the pandas API onSpark?

A. import pyspark.pandas as psdf = ps.DataFrame(spark_df)
B. import pyspark.pandas as psdf = ps.to_pandas(spark_df)
C. spark_df.to_pandas()
D. import pandas as pddf = pd.DataFrame(spark_df)

Show Answer

Question # 8

Which of the following describes the relationship between native Spark DataFrames and pandas APIon Spark DataFrames?

A. pandas API on Spark DataFrames are single-node versions of Spark DataFrames with additionalmetadata
B. pandas API on Spark DataFrames are more performant than Spark DataFrames
C. pandas API on Spark DataFrames are made up of Spark DataFrames and additional metadata
D. pandas API on Spark DataFrames are less mutable versions of Spark DataFrames

Show Answer

Question # 9

Which statement describes a Spark ML transformer?

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

Show Answer

Question # 10

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

A. Keras
B. Scikit-learn
C. PyTorch
D. Spark ML

Show Answer

Question # 11

A data scientist has written a feature engineering notebook that utilizes the pandas library. As thesize of the data processed by the notebook increases, the notebook's runtime is drasticallyincreasing, but it is processing slowly as the size of the data included in the process increases.Which of the following tools can the data scientist use to spend the least amount of time refactoring their notebook to scale with big data?

A. PySpark DataFrame API
B. pandas API on Spark
C. Spark SQL
D. Feature Store

Show Answer

Question # 12

Which of the following hyperparameter optimization methods automatically makes informedselections of hyperparameter values based on previous trials for each iterative model evaluation?

A. Random Search
B. Halving Random Search
C. Tree of Parzen Estimators
D. Grid Search

Show Answer

Question # 13

A data scientist learned during their training to always use 5-fold cross-validation in their modeldevelopment workflow. A colleague suggests that there are cases where a train-validation split couldbe 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 crossvalidationin this scenario?

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

Show Answer

Question # 14

A data scientist is performing hyperparameter tuning using an iterative optimization algorithm. Eachevaluation of unique hyperparameter values is being trained on a single compute node. They areperforming eight total evaluations across eight total compute nodes. While the accuracy of themodel does vary over the eight evaluations, they notice there is no trend of improvement in theaccuracy. The data scientist believes this is due to the parallelization of the tuning process.Which change could the data scientist make to improve their model accuracy over the course of theirtuning process?

A. Change the number of compute nodes to be half or less than half of the number of evaluations.
B. Change the number of compute nodes and the number of evaluations to be much larger butequal.
C. Change the iterative optimization algorithm used to facilitate the tuning process.
D. Change the number of compute nodes to be double or more than double the number ofevaluations.

Show Answer

Question # 15

A data scientist has a Spark DataFrame spark_df. They want to create a new Spark DataFrame thatcontains only the rows from spark_df where the value in column discount is less than or equal 0.Which of the following code blocks will accomplish this task?

A. spark_df.loc[:,spark_df["discount"] <= 0]
B. spark_df[spark_df["discount"] <= 0]
C. spark_df.filter (col("discount") <= 0)
D. spark_df.loc(spark_df["discount"] <= 0, :]

Show Answer

Question # 16

A data scientist has created a linear regression model that uses log(price) as a label variable. Usingthis model, they have performed inference and the predictions and actual label values are in SparkDataFrame preds_df.They are using the following code block to evaluate the model:regression_evaluator.setMetricName("rmse").evaluate(preds_df)Which of the following changes should the data scientist make to evaluate the RMSE in a way that iscomparable with price?

A. They should exponentiate the computed RMSE value
B. They should take the log of the predictions before computing the RMSE
C. They should evaluate the MSE of the log predictions to compute the RMSE
D. They should exponentiate the predictions before computing the RMSE

Show Answer

Question # 17

An organization is developing a feature repository and is electing to one-hot encode all categoricalfeature variables. A data scientist suggests that the categorical feature variables should not be onehotencoded within the feature repository.Which of the following explanations justifies this suggestion?

A. One-hot encoding is a potentially problematic categorical variable strategy for some machinelearning algorithms
B. One-hot encoding is dependent on the target variables values which differ for each apaplication.
C. One-hot encoding is computationally intensive and should only be performed on small samples oftraining sets for individual machine learning problems.
D. One-hot encoding is not a common strategy for representing categorical feature variablesnumerically.

Show Answer

Question # 18

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

A. 3
B. 5
C. 6
D. 18

Show Answer

Question # 19

A data scientist wants to efficiently tune the hyperparameters of a scikit-learn model in parallel. Theyelect to use the Hyperopt library to facilitate this process.Which of the following Hyperopt tools provides the ability to optimize hyperparameters in parallel?

A. fmin
B. SparkTrials
C. quniform
D. search_space
E. objective_function

Show Answer

Question # 20

A data scientist is wanting to explore the Spark DataFrame spark_df. The data scientist wants visualhistograms displaying the distribution of numeric features to be included in the exploration.Which of the following lines of code can the data scientist run to accomplish the task?

A. spark_df.describe()
B. dbutils.data(spark_df).summarize()
C. This task cannot be accomplished in a single line of code.
D. spark_df.summary()
E. dbutils.data.summarize (spark_df)

Show Answer