Available preprocesssing layers
Core preprocesssing layers
TextVectorization layer: turns raw strings into an encoded representation that can be read by an Embedding layer or Dense layer.
Normalization layer: performs feature-wise normalize of input features.
Structured data preprocessing layers
These layers are for structured data encoding and feature engineering.
CategoryEncoding layer: turns integer categorical features into one-hot, multi-hot, or TF-IDF dense representations.
Hashing layer: performs categorical feature hashing, also known as the "hashing trick".
Discretization layer: turns continuous numerical features into integer categorical features.
StringLookup layer: turns string categorical values into integers indices.
IntegerLookup layer: turns integer categorical values into integers indices.
CategoryCrossing layer: combines categorical features into co-occurence features. E.g. if you have feature values "a" and "b", it can provide with the combination feature "a and b are present at the same time".
Image preprocessing layers
These layers are for standardizing the inputs of an image model.
Resizing layer: resizes a batch of images to a target size.
Rescaling layer: rescales and offsets the values of a batch of image (e.g. go from inputs in the [0, 255] range to inputs in the [0, 1] range.
CenterCrop layer: returns a center crop if a batch of images.
Image data augmentation layers
These layers apply random augmentation transforms to a batch of images. They are only active during training.