tf.estimator.Estimator

TensorFlow 1 version View source on GitHub

Estimator class to train and evaluate TensorFlow models.

Inherits From: Estimator

Used in the notebooks

Used in the guide Used in the tutorials

The Estimator object wraps a model which is specified by a model_fn, which, given inputs and a number of other parameters, returns the ops necessary to perform training, evaluation, or predictions.

All outputs (checkpoints, event files, etc.) are written to model_dir, or a subdirectory thereof. If model_dir is not set, a temporary directory is used.

The config argument can be passed tf.estimator.RunConfig object containing information about the execution environment. It is passed on to the model_fn, if the model_fn has a parameter named "config" (and input functions in the same manner). If the config parameter is not passed, it is instantiated by the Estimator. Not passing config means that defaults useful for local execution are used. Estimator makes config available to the model (for instance, to allow specialization based on the number of workers available), and also uses some of its fields to control internals, especially regarding checkpointing.

The params argument contains hyperparameters. It is passed to the model_fn, if the model_fn has a parameter named "params", and to the input functions in the same manner. Estimator only passes params along, it does not inspect it. The structure of params is therefore entirely up to the developer.

None of Estimator's methods can be overridden in subclasses (its constructor enforces this). Subclasses should use model_fn to configure the base class, and may add methods implementing specialized functionality.

See estimators for more information.

To warm-start an Estimator:

estimator = tf.estimator.DNNClassifier(
    feature_columns=[categorical_feature_a_emb, categorical_feature_b_emb],
    hidden_units=[1024, 512, 256],
    warm_start_from="/path/to/checkpoint/dir")

For more details on warm-start configuration, see tf.estimator.WarmStartSettings.

model_fn Model function. Follows the signature:

  • features -- This is the first item returned from the input_fn passed to train, evaluate, and predict. This should be a single tf.Tensor or dict of same.
  • labels -- This is the second item returned from the input_fn passed to train, evaluate, and predict. This should be a single tf.Tensor or dict of same (for multi-head models). If mode is tf.estimator.ModeKeys.PREDICT, labels=None will be passed. If the model_fn's signature does not accept mode, the model_fn must still be able to handle labels=None.
  • mode -- Optional. Specifies if this is training, evaluation or prediction. See tf.estimator.ModeKeys. params -- Optional dict of hyperparameters. Will receive what is passed to Estimator in params parameter. This allows to configure Estimators from hyper parameter tuning.
  • config -- Optional estimator.RunConfig object. Will receive what is passed to Estimator as its config parameter, or a default value. Allows setting up things in your model_fn based on configuration such as num_ps_replicas, or model_dir.
  • Returns -- tf.estimator.EstimatorSpec
model_dir Directory to save model parameters, graph and etc. This can also be used to load checkpoints from the directory into an estimator to continue training a previously saved model. If PathLike object, the path will be resolved. If None, the model_dir in config will be used if set. If both are set, they must be same. If both are None, a temporary directory will be used.
config estimator.RunConfig configuration object.
params dict of hyper parameters that will be passed into model_fn. Keys are names of parameters, values are basic python types.
warm_start_from Optional string filepath to a checkpoint or SavedModel to warm-start from, or a tf.estimator.WarmStartSettings object to fully configure warm-starting. If None, only TRAINABLE variables are warm-started. If the string filepath is provided instead of a tf.estimator.WarmStartSettings, then all variables are warm-started, and it is assumed that vocabularies and tf.Tensor names are unchanged.

ValueError parameters of model_fn don't match params.
ValueError if this is called via a subclass and if that class overrides a member of Estimator.

Eager Compatibility

Calling methods of Estimator will work while eager execution is enabled. However, the model_fn and input_fn is not executed eagerly, Estimator will switch to graph mode before calling all user-provided functions (incl. hooks), so their code has to be compatible with graph mode execution. Note that input_fn code using tf.data generally works in both graph and eager modes.

config

export_savedmodel

model_dir

model_fn Returns the model_fn which is bound to self.params.
params

Methods

eval_dir

View source

Shows the directory name where evaluation metrics are dumped.

Args
name Name of the evaluation if user needs to run multiple evaluations on different data sets, such as on training data vs test data. Metrics for different evaluations are saved in separate folders, and appear separately in tensorboard.

Returns
A string which is the path of directory contains evaluation metrics.

evaluate

View source

Evaluates the model given evaluation data input_fn.

For each step, calls input_fn, which returns one batch of data. Evaluates until:

Args
input_fn A function that constructs the input data for evaluation. See Premade Estimators for more information. The function should construct and return one of the following:

  • A tf.data.Dataset object: Outputs of Dataset object must be a tuple (features, labels) with same constraints as below.
  • A tuple (features, labels): Where features is a tf.Tensor or a dictionary of string feature name to Tensor and labels is a Tensor or a dictionary of string label name to Tensor. Both features and labels are consumed by model_fn. They should satisfy the expectation of model_fn from inputs.
steps Number of steps for which to evaluate model. If None, evaluates until input_fn raises an end-of-input exception.
hooks List of tf.train.SessionRunHook subclass instances. Used for callbacks inside the evaluation call.
checkpoint_path Path of a specific checkpoint to evaluate. If None, the latest checkpoint in model_dir is used. If there are no checkpoints in model_dir, evaluation is run with newly initialized Variables instead of ones restored from checkpoint.
name Name of the evaluation if user needs to run multiple evaluations on different data sets, such as on training data vs test data. Metrics for different evaluations are saved in separate folders, and appear separately in tensorboard.

Returns
A dict containing the evaluation metrics specified in model_fn keyed by name, as well as an entry global_step which contains the value of the global step for which this evaluation was performed. For canned estimators, the dict contains the loss (mean loss per mini-batch) and the average_loss (mean loss per sample). Canned classifiers also return the accuracy. Canned regressors also return the label/mean and the prediction/mean.

Raises
ValueError If steps <= 0.

experimental_export_all_saved_models

View source

Exports a SavedModel with tf.MetaGraphDefs for each requested mode.

For each mode passed in via the input_receiver_fn_map, this method builds a new graph by calling the input_receiver_fn to obtain feature and label Tensors. Next, this method calls the Estimator's model_fn in the passed mode to generate the model graph based on those features and labels, and restores the given checkpoint (or, lacking that, the most recent checkpoint) into the graph. Only one of the modes is used for saving variables to the SavedModel (order of preference: tf.estimator.ModeKeys.TRAIN, tf.estimator.ModeKeys.EVAL, then tf.estimator.ModeKeys.PREDICT), such that up to three tf.MetaGraphDefs are saved with a single set of variables in a single SavedModel directory.

For the variables and tf.MetaGraphDefs, a timestamped export directory below export_dir_base, and writes a SavedModel into it containing the tf.MetaGraphDef for the given mode and its associated signatures.

For prediction, the exported MetaGraphDef will provide one SignatureDef for each element of the export_outputs dict returned from the model_fn, named using the same keys. One of these keys is always tf.saved_model.signature_constants.DEFAULT_SERVING_SIGNATURE_DEF_KEY, indicating which signature will be served when a serving request does not specify one. For each signature, the outputs are provided by the corresponding tf.estimator.export.ExportOutputs, and the inputs are always the input receivers provided by the serving_input_receiver_fn.

For training and evaluation, the train_op is stored in an extra collection, and loss, metrics, and predictions are included in a SignatureDef for the mode in question.

Extra assets may be written into the SavedModel via the assets_extra argument. This should be a dict, where each key gives a destination path (including the filename) relative to the assets.extra directory. The corresponding value gives the full path of the source file to be copied. For example, the simple case of copying a single file without renaming it is specified as {'my_asset_file.txt': '/path/to/my_asset_file.txt'}.