|TensorFlow 1 version||View source on GitHub|
Ops and objects returned from a
model_fn and passed to an
Compat aliases for migration
See Migration guide for more details.
tf.estimator.EstimatorSpec( mode, predictions=None, loss=None, train_op=None, eval_metric_ops=None, export_outputs=None, training_chief_hooks=None, training_hooks=None, scaffold=None, evaluation_hooks=None, prediction_hooks=None )
Used in the notebooks
|Used in the guide||Used in the tutorials|
EstimatorSpec fully defines the model to be run by an
ModeKeys. Specifies if this is training, evaluation or prediction.
Tensoror dict of
loss: Training loss
Tensor. Must be either scalar, or with shape
train_op: Op for the training step.
eval_metric_ops: Dict of metric results keyed by name. The values of the dict can be one of the following: (1) instance of
Metricclass. (2) Results of calling a metric function, namely a
metric_tensorshould be evaluated without any impact on state (typically is a pure computation results based on variables.). For example, it should not trigger the
update_opor requires any input fetching.
export_outputs: Describes the output signatures to be exported to
SavedModeland used during serving. A dict
- name: An arbitrary name for this output.
- output: an
ExportOutputobject such as
PredictOutput. Single-headed models only need to specify one entry in this dictionary. Multi-headed models should specify one entry for each head, one of which must be named using
tf.saved_model.signature_constants.DEFAULT_SERVING_SIGNATURE_DEF_KEY. If no entry is provided, a default
predictionswill be created.
training_chief_hooks: Iterable of
tf.train.SessionRunHookobjects to run on the chief worker during training.
training_hooks: Iterable of
tf.train.SessionRunHookobjects to run on all workers during training.
tf.train.Scaffoldobject that can be used to set initialization, saver, and more to be used in training.
evaluation_hooks: Iterable of
tf.train.SessionRunHookobjects to run during evaluation.
prediction_hooks: Iterable of
tf.train.SessionRunHookobjects to run during predictions.
ValueError: If validation fails.
TypeError: If any of the arguments is not the expected type.