step_categorical_column_with_hash_bucket: Creates a categorical column with hash buckets specification

In tfdatasets: Interface to 'TensorFlow' Datasets

Creates a categorical column with hash buckets specification

Description

Represents sparse feature where ids are set by hashing.

Usage

step_categorical_column_with_hash_bucket(
  spec,
  ...,
  hash_bucket_size,
  dtype = tf$string
)

Arguments

spec	A feature specification created with feature_spec().
...	Comma separated list of variable names to apply the step. selectors can also be used.
hash_bucket_size	An int > 1. The number of buckets.
dtype	The type of features. Only string and integer types are supported.

Value

a FeatureSpec object.

See Also

steps for a complete list of allowed steps.

Other Feature Spec Functions: dataset_use_spec(), feature_spec(), fit.FeatureSpec(), step_bucketized_column(), step_categorical_column_with_identity(), step_categorical_column_with_vocabulary_file(), step_categorical_column_with_vocabulary_list(), step_crossed_column(), step_embedding_column(), step_indicator_column(), step_numeric_column(), step_remove_column(), step_shared_embeddings_column(), steps

Examples

## Not run: 
library(tfdatasets)
data(hearts)
hearts <- tensor_slices_dataset(hearts) %>% dataset_batch(32)

# use the formula interface
spec <- feature_spec(hearts, target ~ thal) %>%
  step_categorical_column_with_hash_bucket(thal, hash_bucket_size = 3)

spec_fit <- fit(spec)
final_dataset <- hearts %>% dataset_use_spec(spec_fit)

## End(Not run)

tfdatasets documentation built on June 30, 2022, 1:04 a.m.