François CholletVerified account

But there are several key advantages of the Functional approach over the subclassing approach: 1. Your model has known inputs shapes. 2. You get access to the internal connectivity graph. 3. The model is a data structure, not a piece of bytecode. Let's see what these are about.

Further, it's even capable of standardizing inputs to what it expects: if you pass data of shape (batch_size,) to a model that expects (batch_size, 1), it will just reshape it. Likewise for dtype conversion (e.g. float64 will get converted to float32).

3. The model is a data structure, not a piece of bytecode. This means it can be cleanly serialized and deserialized -- even across platforms. keras.Model.from_config(functional_model.get_config()) reconstructs the exact same model as the original.

If your model is a Python subclass, to serialize it you could either: a. Pickle the bytecode -- which it completely unsafe, won't work for production, and won't work across platforms

b. Save it as a SavedModel -- which is a form of one-way export (of the TF graph) and won't let you reconstruct the exact same Python object. A graph of layers is a data structure; defining and saving it as a data structure is the intuitive thing to do.