def group_by(self, by=None, **additional_grouping_expressions):
"""
Create an intermediate grouped table expression, pending some group
operation to be applied with it.
Examples
--------
>>> import ibis
>>> pairs = [('a', 'int32'), ('b', 'timestamp'), ('c', 'double')]
>>> t = ibis.table(pairs)
>>> b1, b2 = t.a, t.b
>>> result = t.group_by([b1, b2]).aggregate(sum_of_c=t.c.sum())
Notes
-----
group_by and groupby are equivalent, with `groupby` being provided for
ease-of-use for pandas users.
Returns
-------
grouped_expr : GroupedTableExpr
"""
from ibis.expr.groupby import GroupedTableExpr
return GroupedTableExpr(self, by, **additional_grouping_expressions)
def group_by(self, by):
"""
Create an intermediate grouped table expression, pending some group
operation to be applied with it.
Examples
--------
x.group_by([b1, b2]).aggregate(metrics)
Returns
-------
grouped_expr : GroupedTableExpr
"""
from ibis.expr.groupby import GroupedTableExpr
return GroupedTableExpr(self, by)
def group_by(self, by=None, **additional_grouping_expressions):
"""
Create an intermediate grouped table expression, pending some group
operation to be applied with it.
Examples
--------
x.group_by([b1, b2]).aggregate(metrics)
Notes
-----
group_by and groupby are equivalent, with `groupby` being provided for
ease-of-use for pandas users.
Returns
-------
grouped_expr : GroupedTableExpr
"""
from ibis.expr.groupby import GroupedTableExpr
return GroupedTableExpr(self, by, **additional_grouping_expressions)
def __call__(self, grouped: GroupedTableExpr) -> ir.Expr:
return grouped.aggregate([
operation(grouped.table).name(name)
for name, operation in self.metrics
])
