polars.Expr.list.n_unique#

Expr.list.n_unique() Expr[source]#

Count the number of unique values in every sub-lists.

Examples

>>> df = pl.DataFrame(
...     {
...         "a": [[1, 1, 2], [2, 3, 4]],
...     }
... )
>>> df.with_columns(n_unique=pl.col("a").list.n_unique())
shape: (2, 2)
┌───────────┬──────────┐
│ a         ┆ n_unique │
│ ---       ┆ ---      │
│ list[i64] ┆ u32      │
╞═══════════╪══════════╡
│ [1, 1, 2] ┆ 2        │
│ [2, 3, 4] ┆ 3        │
└───────────┴──────────┘