Feature encoding

The feature encoding determines the way in which example data is turned into numbers that a model can do calculations with. There are five encodings available on the platform: categorical, binary, numeric, image, and text.
Depending on the encoding selected, additional settings and processing is applicable, such as normalization or image transformation.

Types of encoding

Encoding Suitable for Additional settings

Categorical

Text string
Integer

One-hot
Index

Binary

Text string
Integer

Positive class

Numeric

Integer
Float
NumPy array

Normalization

Image

Image

Normalization
Transformation

Text

Text string

Language model
Sequence length