=================================================================================

Some preprocessing layers have an internal state that can be computed based on a sample of the training data. The stateful preprocessing layers are:
          i) TextVectorization. It holds a mapping between string tokens and integer indices
          ii) StringLookup and IntegerLookup. They hold a mapping between input values and integer indices.
          iii) Normalization. It holds the mean and standard deviation of the features.
          iv) Discretization. It holds information about value bucket boundaries.

However, you should avoid using the Keras function adapt() when working with lookup layers with very large vocabularies.

============================================