You are a machine learning specialist for a gaming software startup. Your company is investigating ways to use machine learning to enhance its game software platform. The team has selected the Amazon SageMaker platform for their machine learning efforts. You are participating in the feature transformation process in preparation to create your machine learning models. Instead of transforming your data before you use it in your SageMaker models, you and your team have decided to use the built-in transformations of SageMaker. Specifically, you and your team have decided to use the built-in OneHotEncoder transformer to transform your categorical data.
You have decided to drop one of the categories per feature because you suspect you may have perfectly collinear features. Which of the following is NOT a drop methodology used in the OnHotEncoder transformer?

Answer options:

A.None
B.Last
C.Array
D.First

Answer: B
Option A is incorrect. The OneHotEncoder transformer has the following methodologies you can use to drop one of the categories per feature: None, first, array. None is the default methodology.
Option B is correct. The OneHotEncoder transformer has the following methodologies you can use to drop one of the categories per feature: None, first, array. None is the default methodology. The OneHotEncoder transformer drop parameter does not offer the last methodology.
Option C is incorrect. The OneHotEncoder transformer has the following methodologies you can use to drop one of the categories per feature: None, first, array. None is the default methodology.
Option D is incorrect. The OneHotEncoder transformer has the following methodologies you can use to drop one of the categories per feature: None, first, array. None is the default methodology.
Reference:
Please see the Amazon Machine Learning blog titled Preprocess input data before making predictions using Amazon SageMaker inference pipelines and Scikit-learn, and the Scikit-learn API documentation OneHotEncoder