org.tribuo.interop.tensorflow

Class DenseFeatureConverter

java.lang.Object

org.tribuo.interop.tensorflow.DenseFeatureConverter

All Implemented Interfaces:

com.oracle.labs.mlrg.olcut.config.Configurable, com.oracle.labs.mlrg.olcut.provenance.Provenancable<com.oracle.labs.mlrg.olcut.provenance.ConfiguredObjectProvenance>, Serializable, FeatureConverter

public class DenseFeatureConverter
extends Object
implements FeatureConverter

Converts a sparse example into a dense float vector, then wraps it in a TFloat32.

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
static int	THRESHOLD Feature size beyond which a warning is generated (as TensorFlow requires dense features and large feature spaces are memory hungry).
static int	WARNING_THRESHOLD Number of times the feature size warning should be printed.

Constructor Summary

Constructors

Constructor and Description
DenseFeatureConverter(String inputName) Builds a DenseFeatureConverter, setting the input name.

Method Summary

Modifier and Type	Method and Description
TensorMap	convert(Example<?> example, ImmutableFeatureMap featureIDMap) Converts an Example into a TensorMap suitable for supplying as an input to a graph.
TensorMap	convert(List<? extends Example<?>> examples, ImmutableFeatureMap featureIDMap) Converts a batch of Examples into a single TensorMap suitable for supplying as an input to a graph.
TensorMap	convert(List<? extends SGDVector> vectors) Converts a list of SGDVectors representing a batch of features into a TensorMap.
TensorMap	convert(SGDVector vector) Converts a SGDVector representing the features into a TensorMap.
com.oracle.labs.mlrg.olcut.provenance.ConfiguredObjectProvenance	getProvenance()
Set<String>	inputNamesSet() Gets a view of the names of the inputs this converter produces.
String	toString()

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Methods inherited from interface com.oracle.labs.mlrg.olcut.config.Configurable

postConfig

Field Detail

THRESHOLD

public static final int THRESHOLD

Feature size beyond which a warning is generated (as TensorFlow requires dense features and large feature spaces are memory hungry).

See Also:

Constant Field Values

WARNING_THRESHOLD

public static final int WARNING_THRESHOLD

Number of times the feature size warning should be printed.

See Also:

Constant Field Values

Constructor Detail

DenseFeatureConverter

public DenseFeatureConverter(String inputName)

Builds a DenseFeatureConverter, setting the input name.

Parameters:

inputName - The placeholder input name.

Method Detail

convert

public TensorMap convert(Example<?> example,
                         ImmutableFeatureMap featureIDMap)

Description copied from interface: FeatureConverter

Converts an Example into a TensorMap suitable for supplying as an input to a graph.

It generates it as a single example minibatch.

Specified by:

convert in interface FeatureConverter

Parameters:

example - The example to convert.

featureIDMap - The id map to convert feature names into id numbers.

Returns:

A TensorMap (similar to a TF Python feed_dict) representing the features in this example.

convert

public TensorMap convert(List<? extends Example<?>> examples,
                         ImmutableFeatureMap featureIDMap)

Description copied from interface: FeatureConverter

Converts a batch of Examples into a single TensorMap suitable for supplying as an input to a graph.

Specified by:

convert in interface FeatureConverter

Parameters:

examples - The examples to convert.

featureIDMap - THe id map to convert feature names into id numbers.

Returns:

A TensorMap (similar to a TF Python feed_dict) representing the features in this minibatch.

convert

public TensorMap convert(SGDVector vector)

Description copied from interface: FeatureConverter

Converts a SGDVector representing the features into a TensorMap.

It generates it as a single example minibatch.

Specified by:

convert in interface FeatureConverter

Parameters:

vector - The features to convert.

Returns:

A TensorMap (similar to a TF Python feed_dict) representing this vector.

convert

public TensorMap convert(List<? extends SGDVector> vectors)

Description copied from interface: FeatureConverter

Converts a list of SGDVectors representing a batch of features into a TensorMap.

Specified by:

convert in interface FeatureConverter

Parameters:

vectors - The batch of features to convert.

Returns:

A TensorMap (similar to a TF Python feed_dict) representing this minibatch.

inputNamesSet

public Set<String> inputNamesSet()

Description copied from interface: FeatureConverter

Gets a view of the names of the inputs this converter produces.

Specified by:

inputNamesSet in interface FeatureConverter

Returns:

The input names.

toString

public String toString()

Overrides:

toString in class Object

getProvenance

public com.oracle.labs.mlrg.olcut.provenance.ConfiguredObjectProvenance getProvenance()

Specified by:

getProvenance in interface com.oracle.labs.mlrg.olcut.provenance.Provenancable<com.oracle.labs.mlrg.olcut.provenance.ConfiguredObjectProvenance>