Type DNNLinearCombinedEstimator
Namespace tensorflow.contrib.learn
Parent Estimator
Interfaces IDNNLinearCombinedEstimator
An estimator for TensorFlow Linear and DNN joined training models. THIS CLASS IS DEPRECATED. See
[contrib/learn/README.md](https://www.tensorflow.org/code/tensorflow/contrib/learn/README.md)
for general migration instructions. Note: New users must set `fix_global_step_increment_bug=True` when creating an
estimator. Input of `fit`, `train`, and `evaluate` should have following features,
otherwise there will be a `KeyError`:
if `weight_column_name` is not `None`, a feature with
`key=weight_column_name` whose value is a `Tensor`.
for each `column` in `dnn_feature_columns` + `linear_feature_columns`:
- if `column` is a `SparseColumn`, a feature with `key=column.name`
whose `value` is a `SparseTensor`.
- if `column` is a `WeightedSparseColumn`, two features: the first with
`key` the id column name, the second with `key` the weight column
name. Both features' `value` must be a `SparseTensor`.
- if `column` is a `RealValuedColumn, a feature with `key=column.name`
whose `value` is a `Tensor`.
Methods
Properties
Public static methods
DNNLinearCombinedEstimator NewDyn(object head, object model_dir, object linear_feature_columns, object linear_optimizer, ImplicitContainer<T> _joint_linear_weights, object dnn_feature_columns, object dnn_optimizer, object dnn_hidden_units, object dnn_activation_fn, object dnn_dropout, object gradient_clip_norm, object config, object feature_engineering_fn, object embedding_lr_multipliers, ImplicitContainer<T> fix_global_step_increment_bug, object input_layer_partitioner)
Initializes a DNNLinearCombinedEstimator instance. (deprecated argument values) Warning: SOME ARGUMENT VALUES ARE DEPRECATED: `(fix_global_step_increment_bug=False)`. They will be removed after 2017-04-15.
Instructions for updating:
Please set fix_global_step_increment_bug=True and update training steps in your pipeline. See pydoc for details. Note: New users must set `fix_global_step_increment_bug=True` when creating
an estimator.
Parameters
-
object
head - A _Head object.
-
object
model_dir - Directory to save model parameters, graph and etc. This can also be used to load checkpoints from the directory into a estimator to continue training a previously saved model.
-
object
linear_feature_columns - An iterable containing all the feature columns used by linear part of the model. All items in the set should be instances of classes derived from `FeatureColumn`.
-
object
linear_optimizer - An instance of `tf.Optimizer` used to apply gradients to the linear part of the model. If `None`, will use a FTRL optimizer.
-
ImplicitContainer<T>
_joint_linear_weights - If True will use a single (possibly partitioned) variable to store all weights for the linear model. More efficient if there are many columns, however requires all columns are sparse and have the 'sum' combiner.
-
object
dnn_feature_columns - An iterable containing all the feature columns used by deep part of the model. All items in the set should be instances of classes derived from `FeatureColumn`.
-
object
dnn_optimizer - An instance of `tf.Optimizer` used to apply gradients to the deep part of the model. If `None`, will use an Adagrad optimizer.
-
object
dnn_hidden_units - List of hidden units per layer. All layers are fully connected.
-
object
dnn_activation_fn - Activation function applied to each layer. If `None`,
will use
tf.nn.relu
. -
object
dnn_dropout - When not None, the probability we will drop out a given coordinate.
-
object
gradient_clip_norm - A float > 0. If provided, gradients are clipped to their global norm with this clipping ratio. See tf.clip_by_global_norm for more details.
-
object
config - RunConfig object to configure the runtime settings.
-
object
feature_engineering_fn - Feature engineering function. Takes features and labels which are the output of `input_fn` and returns features and labels which will be fed into the model.
-
object
embedding_lr_multipliers - Optional. A dictionary from `EmbeddingColumn` to a `float` multiplier. Multiplier will be used to multiply with learning rate for the embedding variables.
-
ImplicitContainer<T>
fix_global_step_increment_bug - If `False`, the estimator needs two fit steps to optimize both linear and dnn parts. If `True`, this bug is fixed. New users must set this to `True`, but the default value is `False` for backwards compatibility.
-
object
input_layer_partitioner - Optional. Partitioner for input layer.