R/feature_spec.R

step_remove_column

Creates a step that can remove columns

Description

Removes features of the feature specification.

Usage

 
step_remove_column(spec, ...) 

Arguments

Arguments Description
spec A feature specification created with feature_spec().
Comma separated list of variable names to apply the step. selectors can also be used.

Value

a FeatureSpec object.

Examples

library(tfdatasets) 
data(hearts) 
hearts <- tensor_slices_dataset(hearts) %>% dataset_batch(32) 
 
# use the formula interface 
spec <- feature_spec(hearts, target ~ age) %>% 
  step_numeric_column(age, normalizer_fn = scaler_standard()) %>% 
  step_bucketized_column(age, boundaries = c(20, 50)) %>% 
  step_remove_column(age) 
 
spec_fit <- fit(spec) 
final_dataset <- hearts %>% dataset_use_spec(spec_fit) 

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_hash_bucket(), 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_shared_embeddings_column(), steps