*   Argument `jit_compile` in `Model.compile()` now applies to
        `Model.evaluate()` and `Model.predict()`. Setting `jit_compile=True` in
        `compile()` compiles the model's training, evaluation, and inference
        steps to [XLA](https://www.tensorflow.org/xla). Note that
        `jit_compile=True` may not necessarily work for all models.