Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[REVIEW] Implement scan operations for decimal columns #7707

Merged
merged 3 commits into from
Mar 24, 2021
Merged

[REVIEW] Implement scan operations for decimal columns #7707

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
360b891
implement scan for decimal columns
Mar 24, 2021
78d7c55
Cleanup extra file
Mar 24, 2021
ae77f0c
Move up error
Mar 24, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 3 additions & 0 deletions python/cudf/cudf/core/column/decimal.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -72,6 +72,9 @@ def binary_operator(self, op, other, reflect=False):
result.dtype.precision = _binop_precision(self.dtype, other.dtype, op)
return result

def _apply_scan_op(self, op: str) -> ColumnBase:
return libcudf.reduce.scan(op, self, True)

def as_decimal_column(
self, dtype: Dtype, **kwargs
) -> "cudf.core.column.DecimalColumn":
Expand Down
10 changes: 8 additions & 2 deletions python/cudf/cudf/core/series.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4725,8 +4725,9 @@ def cumsum(self, axis=0, skipna=True, *args, **kwargs):
result_col[first_index:] = None

# pandas always returns int64 dtype if original dtype is int or `bool`
if np.issubdtype(result_col.dtype, np.integer) or np.issubdtype(
result_col.dtype, np.bool_
if not is_decimal_dtype(result_col.dtype) and (
np.issubdtype(result_col.dtype, np.integer)
or np.issubdtype(result_col.dtype, np.bool_)
):
return Series(
result_col.astype(np.int64)._apply_scan_op("sum"),
Expand Down Expand Up @@ -4786,6 +4787,11 @@ def cumprod(self, axis=0, skipna=True, *args, **kwargs):
)
result_col[first_index:] = None

if is_decimal_dtype(result_col.dtype):
galipremsagar marked this conversation as resolved.
Show resolved Hide resolved
raise NotImplementedError(
"cumprod does not currently support decimal types"
)

# pandas always returns int64 dtype if original dtype is int or `bool`
if np.issubdtype(result_col.dtype, np.integer) or np.issubdtype(
result_col.dtype, np.bool_
Expand Down
46 changes: 46 additions & 0 deletions python/cudf/cudf/tests/test_scan.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@

import cudf
from cudf.tests.utils import INTEGER_TYPES, NUMERIC_TYPES, assert_eq, gen_rand
from cudf.core.dtypes import Decimal64Dtype

params_sizes = [0, 1, 2, 5]

Expand Down Expand Up @@ -61,6 +62,21 @@ def test_cumsum_masked():
assert_eq(got, expected)


@pytest.mark.parametrize(
"dtype",
[Decimal64Dtype(8, 4), Decimal64Dtype(10, 5), Decimal64Dtype(12, 7)],
)
def test_cumsum_decimal(dtype):
data = ["243.32", "48.245", "-7234.298", np.nan, "-467.2"]
gser = cudf.Series(data).astype(dtype)
pser = pd.Series(data, dtype="float64")

got = gser.cumsum()
expected = cudf.Series.from_pandas(pser.cumsum()).astype(dtype)

assert_eq(got, expected)


@pytest.mark.parametrize("dtype,nelem", list(_gen_params()))
def test_cummin(dtype, nelem):
if dtype == np.int8:
Expand Down Expand Up @@ -103,6 +119,21 @@ def test_cummin_masked():
assert_eq(gs.cummin(), expected)


@pytest.mark.parametrize(
"dtype",
[Decimal64Dtype(8, 4), Decimal64Dtype(11, 6), Decimal64Dtype(14, 7)],
)
def test_cummin_decimal(dtype):
data = ["8394.294", np.nan, "-9940.444", np.nan, "-23.928"]
gser = cudf.Series(data).astype(dtype)
pser = pd.Series(data, dtype="float64")

got = gser.cummin()
expected = cudf.Series.from_pandas(pser.cummin()).astype(dtype)

assert_eq(got, expected)


@pytest.mark.parametrize("dtype,nelem", list(_gen_params()))
def test_cummax(dtype, nelem):
if dtype == np.int8:
Expand Down Expand Up @@ -145,6 +176,21 @@ def test_cummax_masked():
assert_eq(gs.cummax(), expected)


@pytest.mark.parametrize(
"dtype",
[Decimal64Dtype(8, 4), Decimal64Dtype(11, 6), Decimal64Dtype(14, 7)],
)
def test_cummax_decimal(dtype):
data = [np.nan, "54.203", "8.222", "644.32", "-562.272"]
gser = cudf.Series(data).astype(dtype)
pser = pd.Series(data, dtype="float64")

got = gser.cummax()
expected = cudf.Series.from_pandas(pser.cummax()).astype(dtype)

assert_eq(got, expected)


@pytest.mark.parametrize("dtype,nelem", list(_gen_params()))
def test_cumprod(dtype, nelem):
if dtype == np.int8:
Expand Down