LostTech.TensorFlow : API Documentation

Type tf.keras.estimator

Namespace tensorflow

Methods

Properties

Public static methods

object model_to_estimator(object keras_model, object keras_model_path, object custom_objects, object model_dir, object config, string checkpoint_format)

Constructs an `Estimator` instance from given keras model.

For usage example, please see: [Creating estimators from Keras Models](https://tensorflow.org/guide/estimators#model_to_estimator).

__Sample Weights__ Estimators returned by `model_to_estimator` are configured to handle sample weights (similar to `keras_model.fit(x, y, sample_weights)`). To pass sample weights when training or evaluating the Estimator, the first item returned by the input function should be a dictionary with keys `features` and `sample_weights`. Example below:

``` keras_model = tf.keras.Model(...) keras_model.compile(...)

estimator = tf.keras.estimator.model_to_estimator(keras_model)

def input_fn(): return dataset_ops.Dataset.from_tensors( ({'features': features, 'sample_weights': sample_weights}, targets))

estimator.train(input_fn, steps=1) ```
Parameters
object keras_model
A compiled Keras model object. This argument is mutually exclusive with `keras_model_path`.
object keras_model_path
Path to a compiled Keras model saved on disk, in HDF5 format, which can be generated with the `save()` method of a Keras model. This argument is mutually exclusive with `keras_model`.
object custom_objects
Dictionary for custom objects.
object model_dir
Directory to save `Estimator` model parameters, graph, summary files for TensorBoard, etc.
object config
`RunConfig` to config `Estimator`.
string checkpoint_format
Sets the format of the checkpoint saved by the estimator when training. May be `saver` or `checkpoint`, depending on whether to save checkpoints from tf.train.Saver or tf.train.Checkpoint. This argument currently defaults to `saver`. When 2.0 is released, the default will be `checkpoint`. Estimators use name-based tf.train.Saver checkpoints, while Keras models use object-based checkpoints from tf.train.Checkpoint. Currently, saving object-based checkpoints from `model_to_estimator` is only supported by Functional and Sequential models.
Returns
object
An Estimator from given keras model.

object model_to_estimator_dyn(object keras_model, object keras_model_path, object custom_objects, object model_dir, object config, ImplicitContainer<T> checkpoint_format)

Constructs an `Estimator` instance from given keras model.

For usage example, please see: [Creating estimators from Keras Models](https://tensorflow.org/guide/estimators#model_to_estimator).

__Sample Weights__ Estimators returned by `model_to_estimator` are configured to handle sample weights (similar to `keras_model.fit(x, y, sample_weights)`). To pass sample weights when training or evaluating the Estimator, the first item returned by the input function should be a dictionary with keys `features` and `sample_weights`. Example below:

``` keras_model = tf.keras.Model(...) keras_model.compile(...)

estimator = tf.keras.estimator.model_to_estimator(keras_model)

def input_fn(): return dataset_ops.Dataset.from_tensors( ({'features': features, 'sample_weights': sample_weights}, targets))

estimator.train(input_fn, steps=1) ```
Parameters
object keras_model
A compiled Keras model object. This argument is mutually exclusive with `keras_model_path`.
object keras_model_path
Path to a compiled Keras model saved on disk, in HDF5 format, which can be generated with the `save()` method of a Keras model. This argument is mutually exclusive with `keras_model`.
object custom_objects
Dictionary for custom objects.
object model_dir
Directory to save `Estimator` model parameters, graph, summary files for TensorBoard, etc.
object config
`RunConfig` to config `Estimator`.
ImplicitContainer<T> checkpoint_format
Sets the format of the checkpoint saved by the estimator when training. May be `saver` or `checkpoint`, depending on whether to save checkpoints from tf.train.Saver or tf.train.Checkpoint. This argument currently defaults to `saver`. When 2.0 is released, the default will be `checkpoint`. Estimators use name-based tf.train.Saver checkpoints, while Keras models use object-based checkpoints from tf.train.Checkpoint. Currently, saving object-based checkpoints from `model_to_estimator` is only supported by Functional and Sequential models.
Returns
object
An Estimator from given keras model.

Public properties

PythonFunctionContainer model_to_estimator_fn get;