same implementation as their `tf.losses` equivalent.
*   For Keras model, the individual call of `Model.evaluate` uses no cached data
    for evaluation, while `Model.fit` uses cached data when `validation_data`
    arg is provided for better performance.
*   Adds a `save_traces` argument to `model.save`/ `tf.keras.models.save_model`

                          type: array
                        maxAge:
                          description: Specifies how long the results of a preflight
                            request can be cached.
                          type: string
                      type: object
                    delegate:
                      description: Delegate is used to specify the particular VirtualService

                          type: array
                        maxAge:
                          description: Specifies how long the results of a preflight
                            request can be cached.
                          type: string
                      type: object
                    delegate:
                      description: Delegate is used to specify the particular VirtualService