Oracle 1Z0-1110-26 試験概要:

Oracle 1Z0-1110-26 試験シラバストピック:

Oracle Cloud Infrastructure Data Science Professional 認定 1Z0-1110-26 試験問題：

問題 #1
You have an embarrassingly parallel or distributed batch job on a large amount of data that you consider running using Data Science Jobs. What would be the best approach to run the workload?

A. Create a new job for every job run that you have to run in parallel, because the Data Science Jobs service can have only one job run per job
B. Reconfigure the job run because Data Science Jobs does not support embarrassingly parallel workloads
C. Create the job in Data Science Jobs and then start the number of simultaneous job runs required for your workload
D. Create the job in Data Science Jobs and start a job run. When it is done, start a new job run until you achieve the number of runs required

問題 #2
You are running a pipeline in the OCI Data Science service and want to override some of the pipeline's default settings. Which of the following statements about overriding pipeline defaults is true?

A. Pipeline defaults can be overridden only during pipeline creation.
B. Pipeline defaults cannot be overridden once the pipeline has been created.
C. Pipeline defaults can be overridden before starting the pipeline execution.
D. Pipeline defaults can be overridden only by the Administrator.

問題 #3
As a data scientist for a hardware company, you have been asked to predict the revenue demand for the upcoming quarter. You develop a time series forecasting model to analyze the dat a. Select the correct sequence of steps to predict the revenue demand values for the upcoming quarter.

A. Prepare model, deploy, verify, save, predict
B. Verify, prepare model, deploy, save, predict
C. Prepare model, verify, save, deploy, predict
D. Predict, deploy, save, verify, prepare model

問題 #4
You want to evaluate the relationship between feature values and target variables. You have a large number of observations having a near uniform distribution and the features are highly correlated. Which model explanation technique should you choose?

A. Accumulated Local Effects
B. Feature Permutation Importance Explanations
C. Feature Dependence Explanations
D. Local Interpretable Model-Agnostic Explanations

問題 #5
You realize that your model deployment is about to reach its utilization limit. What would you do to avoid the issue before requests start to fail? Which THREE steps would you perform?

A. Reduce the load balancer bandwidth limit so that fewer requests come in
B. Update the deployment to use fewer instances
C. Delete the deployment
D. Update the deployment to use a larger virtual machine (more CPUs/memory)
E. Update the deployment to add more instances

解説：