Method value_counts

dask/dataframe/dask_expr/_collection.py:4255–4285

(
        self,
        sort=None,
        ascending=False,
        dropna=True,
        normalize=False,
        split_every=None,
        split_out=no_default,
    )

    @derived_from(pd.Series)
    def value_counts(
        self,
        sort=None,
        ascending=False,
        dropna=True,
        normalize=False,
        split_every=None,
        split_out=no_default,
    ):
        if split_out is no_default:
            if isinstance(self.dtype, CategoricalDtype):
                # unobserved or huge categories will lead to oom errors
                if self.cat.known:
                    split_out = 1 + len(self.dtype.categories) // 100_000
                else:
                    split_out = True
            else:
                split_out = True
        if split_out == 1 and split_out is not True and sort is None:
            sort = True

        length = None
        if (split_out > 1 or split_out is True) and normalize:
            frame = self if not dropna else self.dropna()
            length = Len(frame)

        return new_collection(
            ValueCounts(
                self, sort, ascending, dropna, normalize, split_every, split_out, length
            )
        )

    @derived_from(pd.Series)
    def mode(self, dropna=True, split_every=False):