CSV#
A plain-text format that use newlines to delineate rows and a user-specified delimiter (usually a comma) to separate the values of each column in a data table. Comma-Separated Values (CSV) format is a common format but suffers from not being able to retain type information - thus, all data is imported as categorical data (strings) by default.
Note
This implementation of CSV is based on the definition in RFC 4180.
Interfaces: Extractor, Writable
Parameters#
# | Name | Default | Type | Description |
---|---|---|---|---|
1 | path | string | The path to the CSV file. | |
2 | header | false | bool | Does the CSV document have a header as the first row? |
3 | delimiter | ',' | string | The character that delineates the values of the columns of the data table. |
4 | enclosure | '"' | string | The character used to enclose a cell that contains a delimiter in the body. |
Example#
use Rubix\ML\Extractors\CSV;
$extractor = new CSV('example.csv', true, ',', '"');
Additional Methods#
This extractor does not have any additional methods.
References#
-
T. Shafranovich. (2005). Common Format and MIME Type for Comma-Separated Values (CSV) Files. ↩
Last update:
2021-03-27