cudf.core.column.string.StringMethods.character_ngrams

StringMethods.character_ngrams(n: int = 2, as_list: bool = False) → SeriesOrIndex

Generate the n-grams from characters in a column of strings.

Parameters

nint

The degree of the n-gram (number of consecutive characters). Default of 2 for bigrams.

as_listbool

Set to True to return ngrams in a list column where each list element is the ngrams for each string.

Examples

>>> import cudf
>>> str_series = cudf.Series(['abcd','efgh','xyz'])
>>> str_series.str.character_ngrams(2)
0    ab
1    bc
2    cd
3    ef
4    fg
5    gh
6    xy
7    yz
dtype: object
>>> str_series.str.character_ngrams(3)
0    abc
1    bcd
2    efg
3    fgh
4    xyz
dtype: object
>>> str_series.str.character_ngrams(3,True)
0    [abc, bcd]
1    [efg, fgh]
2         [xyz]
dtype: list