train-evaluate-predict: Base Documentation for train, evaluate, and predict.
In tfestimators: Interface to 'TensorFlow' Estimators

Base Documentation for train, evaluate, and predict.

`input_fn`	An input function, typically generated by the `input_fn()` helper function.
`hooks`	A list of R functions, to be used as callbacks inside the training loop. By default, `hook_history_saver(every_n_step = 10)` and `hook_progress_bar()` will be attached if not provided to save the metrics history and create the progress bar.
`checkpoint_path`	The path to a specific model checkpoint to be used for prediction. If `NULL` (the default), the latest checkpoint in `model_dir` is used.