tf.estimator
tf.estimator is a high-level TensorFlow API for creating and training models by encapsulating the functionalities for training, evaluating, predicting and exporting. TensorFlow recently re-branded and released the TF Learn package within TensorFlow under a new name, TF Estimator, probably to avoid confusion with the TFLearn package from tflearn.org.
tf.estimator allows developers to easily extend the package and implement new ML algorithms by using the existing modular components and TensorFlow's low-level APIs, which serve as the building blocks of ML algorithms. Some examples of these building blocks are evaluation metrics, layers, losses, and optimizers.
The main features provided by tf.estimator are described in the next sections.
Estimators
An estimator is a rule that calculates an estimate of a given quantity. Estimators are used to train and evaluate TensorFlow models. Each estimator is an implementation of a particular type of ML algorithm. They currently support...
