Practice Test 2 | Google Cloud Certified Professional Data Engineer | Dumps | Mock Test
You need to build a machine learning model to recognize different animals for a pet shop. The purpose is to scan the photos on their twitter page and get stats about what pets people like sharing while tagging the pet shop brand the most. Due to cost constraints, the project should be as cost-effective as possible, and that includes work hours dedicated to the project.
Which approach will you consider to build the project?
A. Use Cloud ML Engine API and inspect the descriptions returned by the API. Consider the description with highest score.
B. Use Vision API and inspect the descriptions returned by the API. Consider the description with highest score.
C. Use Vision API and inspect the MID values returned by API to recognize the pets in photos.
D. Use Vision API and inspect the descriptions returned by the API. Consider the description with median score.
Answer B.
Google AutoML Vision API automates the training of your own custom machine learning models by simply uploading images and training custom image models with an easy-to-use graphical interface.
Google AutoML Vision is recommended in this scenario because you can build an image recognition model quickly with less work time, comparing to building your very own model from scratch.
Answer A is incorrect: Any approach other than using AutoML vision API is not recommended.
When inspecting returned values from Vision API, here, you need to check the output’s values descriptions to recognize what type of animal recognized by API. Descriptions with highest score should be considered as they have better prediction.
Answer C is incorrect: MID values are not useful for this scenario.
Answer D is incorrect: API does not provide median scores. It provides a rate how likely the description is accurate.
Source(s):
Google Vision API – Detect Labels: https://cloud.google.com/vision/docs/labels
Comments are closed, but trackbacks and pingbacks are open.