QuestionQ54

Prepare data and use feature engineering in Snowflake

A Data Scientist needs to train a supervised machine-learning model on a dataset containing multiple numeric, continuous features. During the exploration phase, the Data Scientist found that the features are approximately normally distributed, with different means and variances across features. The Data Scientist built the pipeline using Snowpark ML.

Which class in snowflake.ml.modeling.preprocessing should the Data Scientist use to produce features with zero mean and unit variance?

  • A StandardScaler
  • B MinMaxScaler
  • C MaxAbsScaler
  • D Normalizer
Explanation

StandardScaler standardizes each feature by removing its mean and scaling it to unit variance. Snowflake documents this behavior for snowflake.ml.modeling.preprocessing.StandardScaler.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!