R/text_line_dataset.R

dataset_decode_delim

Transform a dataset with delimted text lines into a dataset with named columns

Description

Transform a dataset with delimted text lines into a dataset with named columns

Usage

 
dataset_decode_delim(dataset, record_spec, parallel_records = NULL) 

Arguments

Arguments Description
dataset Dataset containing delimited text lines (e.g. a CSV)
record_spec Specification of column names and types (see delim_record_spec()).
parallel_records (Optional) An integer, representing the number of records to decode in parallel. If not specified, records will be processed sequentially.

See Also

Other dataset methods: dataset_batch(), dataset_cache(), dataset_collect(), dataset_concatenate(), dataset_filter(), dataset_interleave(), dataset_map_and_batch(), dataset_map(), dataset_padded_batch(), dataset_prefetch_to_device(), dataset_prefetch(), dataset_reduce(), dataset_repeat(), dataset_shuffle_and_repeat(), dataset_shuffle(), dataset_skip(), dataset_take_while(), dataset_take(), dataset_window()