DeepLearning - Maple Programming Help

Online Help

All Products    Maple    MapleSim


Home : Support : Online Help : Programming : DeepLearning Package : Constructing Feature Columns : DeepLearning/BucketizedColumn

DeepLearning

  

BucketizedColumn

  

bucketized feature column

 

Calling Sequence

Parameters

Description

Details

Examples

Compatibility

Calling Sequence

BucketizedColumn(fc,boundaries)

Parameters

fc

-

feature column

boundaries

-

list of extended_numeric; boundaries for buckets

Description

• 

The BucketizedColumn(fc,boundaries) command creates a new feature column by assigning continuous data represented in fc into a discrete number of buckets defined by boundaries.

• 

This function is part of the DeepLearning package, so it can be used in the short form BucketizedColumn(..) only after executing the command with(DeepLearning). However, it can always be accessed through the long form of the command by using DeepLearning[BucketizedColumn](..).

Details

• 

The implementation of BucketizedColumn uses the tf.feature_column.bucketized_column function from the TensorFlow Python API Documentation. Consult the TensorFlow API documentation for tf.feature_column.bucketized_column for more information.

Examples

Define a feature which takes a single value, in this case a physical measurement from a flower. Then assign it to one of five buckets.

withDeepLearning:

fcNumericColumnPetalLength,shape=1,datatype=float8

fcFeature ColumnNumericColumn(key='PetalLength', shape=(1,), default_value=None, dtype=tf.float64, normalizer_fn=None)

(1)

bcBucketizedColumnfc,2,3.5,5,6.5

bcFeature ColumnBucketizedColumn(source_column=NumericColumn(key='PetalLength', shape=(1,), default_value=None, dtype=tf.float64, normalizer_fn=None), boundaries=(2, 3.5, 5, 6.5))

(2)

Compatibility

• 

The DeepLearning[BucketizedColumn] command was introduced in Maple 2018.

• 

For more information on Maple 2018 changes, see Updates in Maple 2018.

See Also

DeepLearning

DNNClassifier

Feature Column