Skip to content

PERF: statically define classes for is_dtype checks #33364

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

Closed
wants to merge 2 commits into from
Closed

PERF: statically define classes for is_dtype checks #33364

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
ab1947d
perf
TomAugspurger Apr 7, 2020
a178af3
whatsnew
TomAugspurger Apr 7, 2020
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
1 change: 1 addition & 0 deletions doc/source/whatsnew/v1.1.0.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -289,6 +289,7 @@ Performance improvements
sparse values from ``scipy.sparse`` matrices using the
:meth:`DataFrame.sparse.from_spmatrix` constructor (:issue:`32821`,
:issue:`32825`, :issue:`32826`, :issue:`32856`, :issue:`32858`).
- Performance improvement in dtype checking functions in ``pandas.api.types`` (:issue:`33364`)
- Performance improvement in reductions (sum, min, max) for nullable (integer and boolean) dtypes (:issue:`30982`, :issue:`33261`).


Expand Down
66 changes: 35 additions & 31 deletions pandas/core/dtypes/common.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@
Common type operations.
"""

from typing import Any, Callable, Union
from typing import Any, Union
import warnings

import numpy as np
Expand Down Expand Up @@ -199,20 +199,28 @@ def ensure_python_int(value: Union[int, np.integer]) -> int:
return new_value


def classes(*klasses) -> Callable:
""" evaluate if the tipo is a subclass of the klasses """
return lambda tipo: issubclass(tipo, klasses)


def classes_and_not_datetimelike(*klasses) -> Callable:
"""
evaluate if the tipo is a subclass of the klasses
and not a datetimelike
"""
return lambda tipo: (
issubclass(tipo, klasses)
and not issubclass(tipo, (np.datetime64, np.timedelta64))
)
_object_classes = lambda tipo: issubclass(tipo, np.object_)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

better as a dict i would think

_datetime64_classes = lambda tipo: issubclass(tipo, np.datetime64)
_timedelta64_classes = lambda tipo: issubclass(tipo, np.timedelta64)
_any_int_classes = lambda tipo: issubclass(tipo, (np.integer, np.timedelta64))
_int_classes = lambda tipo: issubclass(tipo, np.integer) and not issubclass(
tipo, (np.datetime64, np.timedelta64)
)
_signed_int_classes = lambda tipo: issubclass(
tipo, np.signedinteger
) and not issubclass(tipo, (np.datetime64, np.timedelta64))
_unsigned_int_classes = lambda tipo: issubclass(
tipo, np.unsignedinteger
) and not issubclass(tipo, (np.datetime64, np.timedelta64))
_int64_classes = lambda tipo: issubclass(tipo, np.int64)
_datetime64_or_timedelta64_classes = lambda tipo: issubclass(
tipo, (np.datetime64, np.timedelta64)
)
_numeric_classes = lambda tipo: issubclass(
tipo, (np.number, np.bool_)
) and not issubclass(tipo, (np.datetime64, np.timedelta64))
_float_classes = lambda tipo: issubclass(tipo, np.floating)
_complex_classes = lambda tipo: issubclass(tipo, np.complexfloating)


def is_object_dtype(arr_or_dtype) -> bool:
Expand Down Expand Up @@ -242,7 +250,7 @@ def is_object_dtype(arr_or_dtype) -> bool:
>>> is_object_dtype([1, 2, 3])
False
"""
return _is_dtype_type(arr_or_dtype, classes(np.object_))
return _is_dtype_type(arr_or_dtype, _object_classes)


def is_sparse(arr) -> bool:
Expand Down Expand Up @@ -390,7 +398,7 @@ def is_datetime64_dtype(arr_or_dtype) -> bool:
>>> is_datetime64_dtype([1, 2, 3])
False
"""
return _is_dtype_type(arr_or_dtype, classes(np.datetime64))
return _is_dtype_type(arr_or_dtype, _datetime64_classes)


def is_datetime64tz_dtype(arr_or_dtype) -> bool:
Expand Down Expand Up @@ -457,7 +465,7 @@ def is_timedelta64_dtype(arr_or_dtype) -> bool:
>>> is_timedelta64_dtype('0 days')
False
"""
return _is_dtype_type(arr_or_dtype, classes(np.timedelta64))
return _is_dtype_type(arr_or_dtype, _timedelta64_classes)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

example instead of creating the module level variable I would simply hardcode:
lambda tipo: issubclass(tipo, np.timedelta64) right here. its much more obvious.



def is_period_dtype(arr_or_dtype) -> bool:
Expand Down Expand Up @@ -687,7 +695,7 @@ def is_any_int_dtype(arr_or_dtype) -> bool:
>>> is_any_int_dtype(pd.Index([1, 2.])) # float
False
"""
return _is_dtype_type(arr_or_dtype, classes(np.integer, np.timedelta64))
return _is_dtype_type(arr_or_dtype, _any_int_classes)


def is_integer_dtype(arr_or_dtype) -> bool:
Expand Down Expand Up @@ -741,7 +749,7 @@ def is_integer_dtype(arr_or_dtype) -> bool:
>>> is_integer_dtype(pd.Index([1, 2.])) # float
False
"""
return _is_dtype_type(arr_or_dtype, classes_and_not_datetimelike(np.integer))
return _is_dtype_type(arr_or_dtype, _int_classes)


def is_signed_integer_dtype(arr_or_dtype) -> bool:
Expand Down Expand Up @@ -797,7 +805,7 @@ def is_signed_integer_dtype(arr_or_dtype) -> bool:
>>> is_signed_integer_dtype(np.array([1, 2], dtype=np.uint32)) # unsigned
False
"""
return _is_dtype_type(arr_or_dtype, classes_and_not_datetimelike(np.signedinteger))
return _is_dtype_type(arr_or_dtype, _signed_int_classes)


def is_unsigned_integer_dtype(arr_or_dtype) -> bool:
Expand Down Expand Up @@ -844,9 +852,7 @@ def is_unsigned_integer_dtype(arr_or_dtype) -> bool:
>>> is_unsigned_integer_dtype(np.array([1, 2], dtype=np.uint32))
True
"""
return _is_dtype_type(
arr_or_dtype, classes_and_not_datetimelike(np.unsignedinteger)
)
return _is_dtype_type(arr_or_dtype, _unsigned_int_classes)


def is_int64_dtype(arr_or_dtype) -> bool:
Expand Down Expand Up @@ -896,7 +902,7 @@ def is_int64_dtype(arr_or_dtype) -> bool:
>>> is_int64_dtype(np.array([1, 2], dtype=np.uint32)) # unsigned
False
"""
return _is_dtype_type(arr_or_dtype, classes(np.int64))
return _is_dtype_type(arr_or_dtype, _int64_classes)


def is_datetime64_any_dtype(arr_or_dtype) -> bool:
Expand Down Expand Up @@ -1050,7 +1056,7 @@ def is_datetime_or_timedelta_dtype(arr_or_dtype) -> bool:
>>> is_datetime_or_timedelta_dtype(np.array([], dtype=np.datetime64))
True
"""
return _is_dtype_type(arr_or_dtype, classes(np.datetime64, np.timedelta64))
return _is_dtype_type(arr_or_dtype, _datetime64_or_timedelta64_classes)


def _is_unorderable_exception(e: TypeError) -> bool:
Expand Down Expand Up @@ -1267,9 +1273,7 @@ def is_numeric_dtype(arr_or_dtype) -> bool:
>>> is_numeric_dtype(np.array([], dtype=np.timedelta64))
False
"""
return _is_dtype_type(
arr_or_dtype, classes_and_not_datetimelike(np.number, np.bool_)
)
return _is_dtype_type(arr_or_dtype, _numeric_classes)


def is_string_like_dtype(arr_or_dtype) -> bool:
Expand Down Expand Up @@ -1334,7 +1338,7 @@ def is_float_dtype(arr_or_dtype) -> bool:
>>> is_float_dtype(pd.Index([1, 2.]))
True
"""
return _is_dtype_type(arr_or_dtype, classes(np.floating))
return _is_dtype_type(arr_or_dtype, _float_classes)


def is_bool_dtype(arr_or_dtype) -> bool:
Expand Down Expand Up @@ -1545,7 +1549,7 @@ def is_complex_dtype(arr_or_dtype) -> bool:
>>> is_complex_dtype(np.array([1 + 1j, 5]))
True
"""
return _is_dtype_type(arr_or_dtype, classes(np.complexfloating))
return _is_dtype_type(arr_or_dtype, _complex_classes)


def _is_dtype(arr_or_dtype, condition) -> bool:
Expand Down