AWS Machine Learning Certification

Know your Kinesis Firehose from your Kinesis Data Stream? How about your CNN from your LSTM? Give these new questions a go to check for any knowledge gaps ahead of certifying for the ‘AWS Machine Learning – Specialty’

Created on By adminadmin

AWS Machine Learning – Quiz 3

1 / 9


A data scientist is using SageMaker notebooks for initial development of deep learning models and wishes to be able to test both Tensorflow and MXnet. Which of the following is a valid recommendation?

(Choose One)

2 / 9


A developer is exploring use of AWS Kinesis for a real-time use case.

Which of the following correctly identify Kinesis types &  features?

a) Data Stream: Fully Managed, Storage & Replay, Auto-Scale, Real-Time
b) FireHose: Built-In data transformation, Output to S3, Redshift, Near-Real Time,
c) Data Analytics: Supports streaming ETL, Serverless, Two Machine Learning Algorithms built-in, Schema Discovery
d) FireHose: Data transformation via Lambda, Data Conversion built-in (via Glue), Fully Managed
e) Data Stream: Real-Time, 24 hour default data storage, Manual Capacity provisioning, Serve multiple Apps

(Select one)

3 / 9


Which of the following must be set by the user (required hyperparameters), for SageMaker’s built-in algorithm, Random Cut Forest? (Assume you are running the job through the console)

4 / 9


A developer is using the Sagemaker  Python SDK and has just configured and fit a model.

Which of the following is true regarding completing deployment to a live endpoint using only the SageMaker console?

(Select One)

5 / 9


A data scientist wants to make use of SageMaker’s built-in Linear Learner  implementation’s ability to explore a large range of models and choose the best.

Which of the following  would be required for the algorithm to automatically calibrate and select the best model?


6 / 9


A developer wants to use ‘R’ to train and deploy models in SageMaker. Which of the following are true?

7 / 9


Which of the following are true regarding SageMaker’s built-in topic modelling algorithms?

a) NTM and LDA are both supervised topic modelling algorithms and available as built-in algorithms in SageMaker.
b) LDA has many more hyperparameters than NTM allowing greater tuning.
c) NTM provides an auxiliary input channel,  to make top list of topic words more easily accessible.
d)Both NTM and LDA support pipe mode only.

8 / 9


A social running & cycling app Product Owner wants to provide its users with event recommendations based on users previous ratings and participations.

Which of the following could be used to build a recommendation service?

(Select Two)

9 / 9


Which of the following must be set by the user (required hyperparameters), for SageMaker’s built-in algorithm, Factorization Machines? (assume regression)

Your score is

The average score is 36%


No responses yet

    Leave a Reply

    Your email address will not be published. Required fields are marked *