One Hot Encoder#
The One Hot Encoder takes a feature column of categorical values and produces an n-d one-hot representation where n is equal to the number of unique categories in that column. After the transformation, a 0 in any location indicates that the category represented by that column is not present in the sample whereas a 1 indicates that a category is present. One hot encoding is typically used to convert categorical data to continuous so that it can be used to train a learner that is only compatible with continuous features.
Data Type Compatibility: Categorical
This transformer does not have any parameters.
Return the categories computed during fitting indexed by feature column:
public categories() : ?array
use Rubix\ML\Transformers\OneHotEncoder; $transformer = new OneHotEncoder();