Use this when each of your sparse inputs has both an ID and a value. For example, if you're representing text documents as a collection of word frequencies, you can provide 2 parallel sparse input features ('terms' and 'frequencies' below).
column_categorical_weighted(categorical_column, weight_feature_key,
dtype = tf$float32)
A categorical column created by
column_categorical_*()
functions.
String key for weight values.
Type of weights, such as tf$float32
. Only float and integer
weights are supported.
A categorical column composed of two sparse features: one represents id, the other represents weight (value) of the id feature in that example.
ValueError: if dtype
is not convertible to float.
Other feature column constructors: column_bucketized
,
column_categorical_with_hash_bucket
,
column_categorical_with_identity
,
column_categorical_with_vocabulary_file
,
column_categorical_with_vocabulary_list
,
column_crossed
,
column_embedding
,
column_numeric
, input_layer