Module: tf.estimator

Estimator: High level tools for working with models.

Modules

experimental module: Public API for tf.estimator.experimental namespace.

export module: All public utility methods for exporting Estimator to SavedModel.

Classes

class BaselineClassifier: A classifier that can establish a simple baseline. (deprecated)

class BaselineEstimator: An estimator that can establish a simple baseline. (deprecated)

class BaselineRegressor: A regressor that can establish a simple baseline. (deprecated)

class BestExporter: This class exports the serving graph and checkpoints of the best models. (deprecated)

class BinaryClassHead: Creates a Head for single label binary classification. (deprecated)

class CheckpointSaverHook: Saves checkpoints every N steps or seconds. (deprecated)

class CheckpointSaverListener: Interface for listeners that take action before or after checkpoint save. (deprecated)

class DNNClassifier: A classifier for TensorFlow DNN models. (deprecated)

class DNNEstimator: An estimator for TensorFlow DNN models with user-specified head. (deprecated)

class DNNLinearCombinedClassifier: An estimator for TensorFlow Linear and DNN joined classification models. (deprecated)

class DNNLinearCombinedEstimator: An estimator for TensorFlow Linear and DNN joined models with custom head. (deprecated)

class DNNLinearCombinedRegressor: An estimator for TensorFlow Linear and DNN joined models for regression. (deprecated)

class DNNRegressor: A regressor for TensorFlow DNN models. (deprecated)

class Estimator: Estimator class to train and evaluate TensorFlow models. (deprecated)

class EstimatorSpec: Ops and objects returned from a model_fn and passed to an Estimator. (deprecated)

class EvalSpec: Configuration for the "eval" part for the train_and_evaluate call. (deprecated)

class Exporter: A class representing a type of model export. (deprecated)

class FeedFnHook: Runs feed_fn and sets the feed_dict accordingly. (deprecated)

class FinalExporter: This class exports the serving graph and checkpoints at the end. (deprecated)

class FinalOpsHook: A hook which evaluates Tensors at the end of a session. (deprecated)

class GlobalStepWaiterHook: Delays execution until global step reaches wait_until_step. (deprecated)

class Head: Interface for the head/top of a model. (deprecated)

class LatestExporter: This class regularly exports the serving graph and checkpoints. (deprecated)

class LinearClassifier: Linear classifier model. (deprecated)

class LinearEstimator: An estimator for TensorFlow linear models with user-specified head. (deprecated)

class LinearRegressor: An estimator for TensorFlow Linear regression problems. (deprecated)

class LoggingTensorHook: Prints the given tensors every N local steps, every N seconds, or at end. (deprecated)

class LogisticRegressionHead: Creates a Head for logistic regression. (deprecated)

class ModeKeys: Standard names for Estimator model modes. (deprecated)

class MultiClassHead: Creates a Head for multi class classification. (deprecated)

class MultiHead: Creates a Head for multi-objective learning. (deprecated)

class MultiLabelHead: Creates a Head for multi-label classification. (deprecated)

class NanLossDuringTrainingError: DEPRECATED FUNCTION

class NanTensorHook: Monitors the loss tensor and stops training if loss is NaN. (deprecated)

class PoissonRegressionHead: Creates a Head for poisson regression using tf.nn.log_poisson_loss. (deprecated)

class ProfilerHook: Captures CPU/GPU profiling information every N steps or seconds. (deprecated)

class RegressionHead: Creates a Head for regression using the mean_squared_error loss. (deprecated)

class RunConfig: This class specifies the configurations for an Estimator run. (deprecated)

class SecondOrStepTimer: Timer that triggers at most once every N seconds or once every N steps. (deprecated)

class SessionRunArgs: Represents arguments to be added to a Session.run() call. (deprecated)

class SessionRunContext: Provides information about the session.run() call being made. (deprecated)

class SessionRunHook: Hook to extend calls to MonitoredSession.run(). (deprecated)

class SessionRunValues: Contains the results of Session.run(). (deprecated)

class StepCounterHook: Hook that counts steps per second. (deprecated)

class StopAtStepHook: Hook that requests stop at a specified step. (deprecated)

class SummarySaverHook: Saves summaries every N steps. (deprecated)

class TrainSpec: Configuration for the "train" part for the train_and_evaluate call. (deprecated)

class VocabInfo: Vocabulary information for warm-starting. (deprecated)

class WarmStartSettings: Settings for warm-starting in tf.estimator.Estimators. (deprecated)

Functions

add_metrics(...): Creates a new tf.estimator.Estimator which has given metrics. (deprecated)

classifier_parse_example_spec(...): Generates parsing spec for tf.parse_example to be used with classifiers. (deprecated)

regressor_parse_example_spec(...): Generates parsing spec for tf.parse_example to be used with regressors. (deprecated)

train_and_evaluate(...): Train and evaluate the estimator. (deprecated)