def _is_type_compatible(a, b):
"""helper for interval_range to check type compat of start/end/freq"""
is_ts_compat = lambda x: isinstance(x, (Timestamp, DateOffset))
is_td_compat = lambda x: isinstance(x, (Timedelta, DateOffset))
return ((is_number(a) and is_number(b)) or
(is_ts_compat(a) and is_ts_compat(b)) or
(is_td_compat(a) and is_td_compat(b)) or
com._any_none(a, b))
def _is_type_compatible(a, b) -> bool:
"""
Helper for interval_range to check type compat of start/end/freq.
"""
is_ts_compat = lambda x: isinstance(x, (Timestamp, BaseOffset))
is_td_compat = lambda x: isinstance(x, (Timedelta, BaseOffset))
return (
(is_number(a) and is_number(b))
or (is_ts_compat(a) and is_ts_compat(b))
or (is_td_compat(a) and is_td_compat(b))
or com.any_none(a, b)
)
def _validate_fill_value(self, value):
if is_bool(value) or is_bool_dtype(value):
# force conversion to object
# so we don't lose the bools
raise TypeError
elif is_scalar(value) and isna(value):
if is_valid_nat_for_dtype(value, self.dtype):
value = self._na_value
if self.dtype.kind != "f":
# raise so that caller can cast
raise TypeError
else:
# NaT, np.datetime64("NaT"), np.timedelta64("NaT")
raise TypeError
elif is_scalar(value):
if not is_number(value):
# e.g. datetime64, timedelta64, datetime, ...
raise TypeError
elif lib.is_complex(value):
# at least until we have a ComplexIndx
raise TypeError
elif is_float(value) and self.dtype.kind != "f":
if not value.is_integer():
raise TypeError
value = int(value)
return value
def _validate_fill_value(self, value):
if is_bool(value) or is_bool_dtype(value):
# force conversion to object
# so we don't lose the bools
raise TypeError
elif is_scalar(value) and isna(value):
if is_valid_nat_for_dtype(value, self.dtype):
value = self._na_value
if self.dtype.kind != "f":
# raise so that caller can cast
raise TypeError
else:
# NaT, np.datetime64("NaT"), np.timedelta64("NaT")
raise TypeError
elif is_scalar(value):
if not is_number(value):
# e.g. datetime64, timedelta64, datetime, ...
raise TypeError
elif lib.is_complex(value):
# at least until we have a ComplexIndx
raise TypeError
elif is_float(value) and self.dtype.kind != "f":
if not value.is_integer():
raise TypeError
value = int(value)
elif hasattr(value, "dtype") and value.dtype.kind in ["m", "M"]:
# TODO: if we're checking arraylike here, do so systematically
raise TypeError
return value
def _is_valid_endpoint(endpoint):
"""helper for interval_range to check if start/end are valid types"""
return any([
is_number(endpoint),
isinstance(endpoint, Timestamp),
isinstance(endpoint, Timedelta), endpoint is None
])
def assert_attr_equal(attr: str, left, right, obj: str = "Attributes"):
"""
Check attributes are equal. Both objects must have attribute.
Parameters
----------
attr : str
Attribute name being compared.
left : object
right : object
obj : str, default 'Attributes'
Specify object name being compared, internally used to show appropriate
assertion message
"""
__tracebackhide__ = True
left_attr = getattr(left, attr)
right_attr = getattr(right, attr)
if left_attr is right_attr:
return True
elif (is_number(left_attr) and np.isnan(left_attr)
and is_number(right_attr) and np.isnan(right_attr)):
# np.nan
return True
elif (isinstance(left_attr, (np.datetime64, np.timedelta64))
and isinstance(right_attr, (np.datetime64, np.timedelta64))
and type(left_attr) is type(right_attr) and np.isnat(left_attr)
and np.isnat(right_attr)):
# np.datetime64("nat") or np.timedelta64("nat")
return True
try:
result = left_attr == right_attr
except TypeError:
# datetimetz on rhs may raise TypeError
result = False
if (left_attr is pd.NA) ^ (right_attr is pd.NA):
result = False
elif not isinstance(result, bool):
result = result.all()
if result:
return True
else:
msg = f'Attribute "{attr}" are different'
raise_assert_detail(obj, msg, left_attr, right_attr)
def test_agg_cython_table(self, series, func, expected):
# GH21224
# test reducing functions in
# pandas.core.base.SelectionMixin._cython_table
result = series.agg(func)
if is_number(expected):
assert np.isclose(result, expected, equal_nan=True)
else:
assert result == expected
def _check_as_is(x):
return (self.quoting == csv.QUOTE_NONNUMERIC and
is_number(x)) or isinstance(x, str)
def test_is_number(self):
assert is_number(True)
assert is_number(1)
assert is_number(1.1)
assert is_number(1 + 3j)
assert is_number(np.bool(False))
assert is_number(np.int64(1))
assert is_number(np.float64(1.1))
assert is_number(np.complex128(1 + 3j))
assert is_number(np.nan)
assert not is_number(None)
assert not is_number('x')
assert not is_number(datetime(2011, 1, 1))
assert not is_number(np.datetime64('2011-01-01'))
assert not is_number(Timestamp('2011-01-01'))
assert not is_number(Timestamp('2011-01-01', tz='US/Eastern'))
assert not is_number(timedelta(1000))
assert not is_number(Timedelta('1 days'))
# questionable
assert not is_number(np.bool_(False))
assert is_number(np.timedelta64(1, 'D'))
def _is_numeric(x):
""" Whether the array x is numeric. """
return all(is_number(i) for i in x)
def _check_as_is(x):
return (self.quoting == csv.QUOTE_NONNUMERIC and
is_number(x)) or isinstance(x, str)
def interval_range(start=None, end=None, periods=None, freq=None,
name=None, closed='right'):
"""
Return a fixed frequency IntervalIndex
Parameters
----------
start : numeric or datetime-like, default None
Left bound for generating intervals
end : numeric or datetime-like, default None
Right bound for generating intervals
periods : integer, default None
Number of periods to generate
freq : numeric, string, or DateOffset, default None
The length of each interval. Must be consistent with the type of start
and end, e.g. 2 for numeric, or '5H' for datetime-like. Default is 1
for numeric and 'D' (calendar daily) for datetime-like.
name : string, default None
Name of the resulting IntervalIndex
closed : string, default 'right'
options are: 'left', 'right', 'both', 'neither'
Notes
-----
Of the three parameters: ``start``, ``end``, and ``periods``, exactly two
must be specified.
Returns
-------
rng : IntervalIndex
Examples
--------
Numeric ``start`` and ``end`` is supported.
>>> pd.interval_range(start=0, end=5)
IntervalIndex([(0, 1], (1, 2], (2, 3], (3, 4], (4, 5]]
closed='right', dtype='interval[int64]')
Additionally, datetime-like input is also supported.
>>> pd.interval_range(start=pd.Timestamp('2017-01-01'),
end=pd.Timestamp('2017-01-04'))
IntervalIndex([(2017-01-01, 2017-01-02], (2017-01-02, 2017-01-03],
(2017-01-03, 2017-01-04]]
closed='right', dtype='interval[datetime64[ns]]')
The ``freq`` parameter specifies the frequency between the left and right.
endpoints of the individual intervals within the ``IntervalIndex``. For
numeric ``start`` and ``end``, the frequency must also be numeric.
>>> pd.interval_range(start=0, periods=4, freq=1.5)
IntervalIndex([(0.0, 1.5], (1.5, 3.0], (3.0, 4.5], (4.5, 6.0]]
closed='right', dtype='interval[float64]')
Similarly, for datetime-like ``start`` and ``end``, the frequency must be
convertible to a DateOffset.
>>> pd.interval_range(start=pd.Timestamp('2017-01-01'),
periods=3, freq='MS')
IntervalIndex([(2017-01-01, 2017-02-01], (2017-02-01, 2017-03-01],
(2017-03-01, 2017-04-01]]
closed='right', dtype='interval[datetime64[ns]]')
The ``closed`` parameter specifies which endpoints of the individual
intervals within the ``IntervalIndex`` are closed.
>>> pd.interval_range(end=5, periods=4, closed='both')
IntervalIndex([[1, 2], [2, 3], [3, 4], [4, 5]]
closed='both', dtype='interval[int64]')
See Also
--------
IntervalIndex : an Index of intervals that are all closed on the same side.
"""
if com._count_not_none(start, end, periods) != 2:
raise ValueError('Of the three parameters: start, end, and periods, '
'exactly two must be specified')
start = com._maybe_box_datetimelike(start)
end = com._maybe_box_datetimelike(end)
endpoint = next(com._not_none(start, end))
if not _is_valid_endpoint(start):
msg = 'start must be numeric or datetime-like, got {start}'
raise ValueError(msg.format(start=start))
if not _is_valid_endpoint(end):
msg = 'end must be numeric or datetime-like, got {end}'
raise ValueError(msg.format(end=end))
if is_float(periods):
periods = int(periods)
elif not is_integer(periods) and periods is not None:
msg = 'periods must be a number, got {periods}'
raise TypeError(msg.format(periods=periods))
freq = freq or (1 if is_number(endpoint) else 'D')
if not is_number(freq):
try:
freq = to_offset(freq)
except ValueError:
raise ValueError('freq must be numeric or convertible to '
'DateOffset, got {freq}'.format(freq=freq))
# verify type compatibility
if not all([_is_type_compatible(start, end),
_is_type_compatible(start, freq),
_is_type_compatible(end, freq)]):
raise TypeError("start, end, freq need to be type compatible")
if is_number(endpoint):
if periods is None:
periods = int((end - start) // freq)
if start is None:
start = end - periods * freq
# force end to be consistent with freq (lower if freq skips over end)
end = start + periods * freq
# end + freq for inclusive endpoint
breaks = np.arange(start, end + freq, freq)
elif isinstance(endpoint, Timestamp):
# add one to account for interval endpoints (n breaks = n-1 intervals)
if periods is not None:
periods += 1
breaks = date_range(start=start, end=end, periods=periods, freq=freq)
else:
# add one to account for interval endpoints (n breaks = n-1 intervals)
if periods is not None:
periods += 1
breaks = timedelta_range(start=start, end=end, periods=periods,
freq=freq)
return IntervalIndex.from_breaks(breaks, name=name, closed=closed)
def _is_numeric(x):
""" Whether the array x is numeric. """
print("x is: ",x)
return all(is_number(i) for i in x)
def interval_range(
start=None,
end=None,
periods=None,
freq=None,
name: Hashable = None,
closed: lib.NoDefault = lib.no_default,
inclusive: IntervalClosedType | None = None,
) -> IntervalIndex:
"""
Return a fixed frequency IntervalIndex.
Parameters
----------
start : numeric or datetime-like, default None
Left bound for generating intervals.
end : numeric or datetime-like, default None
Right bound for generating intervals.
periods : int, default None
Number of periods to generate.
freq : numeric, str, or DateOffset, default None
The length of each interval. Must be consistent with the type of start
and end, e.g. 2 for numeric, or '5H' for datetime-like. Default is 1
for numeric and 'D' for datetime-like.
name : str, default None
Name of the resulting IntervalIndex.
closed : {'left', 'right', 'both', 'neither'}, default 'right'
Whether the intervals are closed on the left-side, right-side, both
or neither.
.. deprecated:: 1.5.0
Argument `closed` has been deprecated to standardize boundary inputs.
Use `inclusive` instead, to set each bound as closed or open.
inclusive : {"both", "neither", "left", "right"}, default "both"
Include boundaries; Whether to set each bound as closed or open.
.. versionadded:: 1.5.0
Returns
-------
IntervalIndex
See Also
--------
IntervalIndex : An Index of intervals that are all closed on the same side.
Notes
-----
Of the four parameters ``start``, ``end``, ``periods``, and ``freq``,
exactly three must be specified. If ``freq`` is omitted, the resulting
``IntervalIndex`` will have ``periods`` linearly spaced elements between
``start`` and ``end``, inclusively.
To learn more about datetime-like frequency strings, please see `this link
<https://pandas.pydata.org/pandas-docs/stable/user_guide/timeseries.html#offset-aliases>`__.
Examples
--------
Numeric ``start`` and ``end`` is supported.
>>> pd.interval_range(start=0, end=5, inclusive="right")
IntervalIndex([(0, 1], (1, 2], (2, 3], (3, 4], (4, 5]],
dtype='interval[int64, right]')
Additionally, datetime-like input is also supported.
>>> pd.interval_range(start=pd.Timestamp('2017-01-01'),
... end=pd.Timestamp('2017-01-04'), inclusive="right")
IntervalIndex([(2017-01-01, 2017-01-02], (2017-01-02, 2017-01-03],
(2017-01-03, 2017-01-04]],
dtype='interval[datetime64[ns], right]')
The ``freq`` parameter specifies the frequency between the left and right.
endpoints of the individual intervals within the ``IntervalIndex``. For
numeric ``start`` and ``end``, the frequency must also be numeric.
>>> pd.interval_range(start=0, periods=4, freq=1.5, inclusive="right")
IntervalIndex([(0.0, 1.5], (1.5, 3.0], (3.0, 4.5], (4.5, 6.0]],
dtype='interval[float64, right]')
Similarly, for datetime-like ``start`` and ``end``, the frequency must be
convertible to a DateOffset.
>>> pd.interval_range(start=pd.Timestamp('2017-01-01'),
... periods=3, freq='MS', inclusive="right")
IntervalIndex([(2017-01-01, 2017-02-01], (2017-02-01, 2017-03-01],
(2017-03-01, 2017-04-01]],
dtype='interval[datetime64[ns], right]')
Specify ``start``, ``end``, and ``periods``; the frequency is generated
automatically (linearly spaced).
>>> pd.interval_range(start=0, end=6, periods=4, inclusive="right")
IntervalIndex([(0.0, 1.5], (1.5, 3.0], (3.0, 4.5], (4.5, 6.0]],
dtype='interval[float64, right]')
The ``inclusive`` parameter specifies which endpoints of the individual
intervals within the ``IntervalIndex`` are closed.
>>> pd.interval_range(end=5, periods=4, inclusive='both')
IntervalIndex([[1, 2], [2, 3], [3, 4], [4, 5]],
dtype='interval[int64, both]')
"""
if inclusive is not None and not isinstance(closed, lib.NoDefault):
raise ValueError(
"Deprecated argument `closed` cannot be passed "
"if argument `inclusive` is not None"
)
elif not isinstance(closed, lib.NoDefault):
warnings.warn(
"Argument `closed` is deprecated in favor of `inclusive`.",
FutureWarning,
stacklevel=2,
)
if closed is None:
inclusive = "both"
elif closed in ("both", "neither", "left", "right"):
inclusive = closed
else:
raise ValueError(
"Argument `closed` has to be either"
"'both', 'neither', 'left' or 'right'"
)
elif inclusive is None:
inclusive = "both"
start = maybe_box_datetimelike(start)
end = maybe_box_datetimelike(end)
endpoint = start if start is not None else end
if freq is None and com.any_none(periods, start, end):
freq = 1 if is_number(endpoint) else "D"
if com.count_not_none(start, end, periods, freq) != 3:
raise ValueError(
"Of the four parameters: start, end, periods, and "
"freq, exactly three must be specified"
)
if not _is_valid_endpoint(start):
raise ValueError(f"start must be numeric or datetime-like, got {start}")
elif not _is_valid_endpoint(end):
raise ValueError(f"end must be numeric or datetime-like, got {end}")
if is_float(periods):
periods = int(periods)
elif not is_integer(periods) and periods is not None:
raise TypeError(f"periods must be a number, got {periods}")
if freq is not None and not is_number(freq):
try:
freq = to_offset(freq)
except ValueError as err:
raise ValueError(
f"freq must be numeric or convertible to DateOffset, got {freq}"
) from err
# verify type compatibility
if not all(
[
_is_type_compatible(start, end),
_is_type_compatible(start, freq),
_is_type_compatible(end, freq),
]
):
raise TypeError("start, end, freq need to be type compatible")
# +1 to convert interval count to breaks count (n breaks = n-1 intervals)
if periods is not None:
periods += 1
breaks: np.ndarray | TimedeltaIndex | DatetimeIndex
if is_number(endpoint):
# force consistency between start/end/freq (lower end if freq skips it)
if com.all_not_none(start, end, freq):
end -= (end - start) % freq
# compute the period/start/end if unspecified (at most one)
if periods is None:
periods = int((end - start) // freq) + 1
elif start is None:
start = end - (periods - 1) * freq
elif end is None:
end = start + (periods - 1) * freq
breaks = np.linspace(start, end, periods)
if all(is_integer(x) for x in com.not_none(start, end, freq)):
# np.linspace always produces float output
# error: Incompatible types in assignment (expression has type
# "Union[ExtensionArray, ndarray]", variable has type "ndarray")
breaks = maybe_downcast_numeric( # type: ignore[assignment]
breaks, np.dtype("int64")
)
else:
# delegate to the appropriate range function
if isinstance(endpoint, Timestamp):
breaks = date_range(start=start, end=end, periods=periods, freq=freq)
else:
breaks = timedelta_range(start=start, end=end, periods=periods, freq=freq)
return IntervalIndex.from_breaks(breaks, name=name, closed=inclusive)
def interval_range(
start=None, end=None, periods=None, freq=None, name=None, closed="right"
):
"""
Return a fixed frequency IntervalIndex.
Parameters
----------
start : numeric or datetime-like, default None
Left bound for generating intervals.
end : numeric or datetime-like, default None
Right bound for generating intervals.
periods : int, default None
Number of periods to generate.
freq : numeric, str, or DateOffset, default None
The length of each interval. Must be consistent with the type of start
and end, e.g. 2 for numeric, or '5H' for datetime-like. Default is 1
for numeric and 'D' for datetime-like.
name : str, default None
Name of the resulting IntervalIndex.
closed : {'left', 'right', 'both', 'neither'}, default 'right'
Whether the intervals are closed on the left-side, right-side, both
or neither.
Returns
-------
IntervalIndex
See Also
--------
IntervalIndex : An Index of intervals that are all closed on the same side.
Notes
-----
Of the four parameters ``start``, ``end``, ``periods``, and ``freq``,
exactly three must be specified. If ``freq`` is omitted, the resulting
``IntervalIndex`` will have ``periods`` linearly spaced elements between
``start`` and ``end``, inclusively.
To learn more about datetime-like frequency strings, please see `this link
<https://pandas.pydata.org/pandas-docs/stable/user_guide/timeseries.html#offset-aliases>`__.
Examples
--------
Numeric ``start`` and ``end`` is supported.
>>> pd.interval_range(start=0, end=5)
IntervalIndex([(0, 1], (1, 2], (2, 3], (3, 4], (4, 5]],
closed='right', dtype='interval[int64]')
Additionally, datetime-like input is also supported.
>>> pd.interval_range(start=pd.Timestamp('2017-01-01'),
... end=pd.Timestamp('2017-01-04'))
IntervalIndex([(2017-01-01, 2017-01-02], (2017-01-02, 2017-01-03],
(2017-01-03, 2017-01-04]],
closed='right', dtype='interval[datetime64[ns]]')
The ``freq`` parameter specifies the frequency between the left and right.
endpoints of the individual intervals within the ``IntervalIndex``. For
numeric ``start`` and ``end``, the frequency must also be numeric.
>>> pd.interval_range(start=0, periods=4, freq=1.5)
IntervalIndex([(0.0, 1.5], (1.5, 3.0], (3.0, 4.5], (4.5, 6.0]],
closed='right', dtype='interval[float64]')
Similarly, for datetime-like ``start`` and ``end``, the frequency must be
convertible to a DateOffset.
>>> pd.interval_range(start=pd.Timestamp('2017-01-01'),
... periods=3, freq='MS')
IntervalIndex([(2017-01-01, 2017-02-01], (2017-02-01, 2017-03-01],
(2017-03-01, 2017-04-01]],
closed='right', dtype='interval[datetime64[ns]]')
Specify ``start``, ``end``, and ``periods``; the frequency is generated
automatically (linearly spaced).
>>> pd.interval_range(start=0, end=6, periods=4)
IntervalIndex([(0.0, 1.5], (1.5, 3.0], (3.0, 4.5], (4.5, 6.0]],
closed='right',
dtype='interval[float64]')
The ``closed`` parameter specifies which endpoints of the individual
intervals within the ``IntervalIndex`` are closed.
>>> pd.interval_range(end=5, periods=4, closed='both')
IntervalIndex([[1, 2], [2, 3], [3, 4], [4, 5]],
closed='both', dtype='interval[int64]')
"""
start = com.maybe_box_datetimelike(start)
end = com.maybe_box_datetimelike(end)
endpoint = start if start is not None else end
if freq is None and com.any_none(periods, start, end):
freq = 1 if is_number(endpoint) else "D"
if com.count_not_none(start, end, periods, freq) != 3:
raise ValueError(
"Of the four parameters: start, end, periods, and "
"freq, exactly three must be specified"
)
if not _is_valid_endpoint(start):
raise ValueError(f"start must be numeric or datetime-like, got {start}")
elif not _is_valid_endpoint(end):
raise ValueError(f"end must be numeric or datetime-like, got {end}")
if is_float(periods):
periods = int(periods)
elif not is_integer(periods) and periods is not None:
raise TypeError(f"periods must be a number, got {periods}")
if freq is not None and not is_number(freq):
try:
freq = to_offset(freq)
except ValueError as err:
raise ValueError(
f"freq must be numeric or convertible to DateOffset, got {freq}"
) from err
# verify type compatibility
if not all(
[
_is_type_compatible(start, end),
_is_type_compatible(start, freq),
_is_type_compatible(end, freq),
]
):
raise TypeError("start, end, freq need to be type compatible")
# +1 to convert interval count to breaks count (n breaks = n-1 intervals)
if periods is not None:
periods += 1
if is_number(endpoint):
# force consistency between start/end/freq (lower end if freq skips it)
if com.all_not_none(start, end, freq):
end -= (end - start) % freq
# compute the period/start/end if unspecified (at most one)
if periods is None:
periods = int((end - start) // freq) + 1
elif start is None:
start = end - (periods - 1) * freq
elif end is None:
end = start + (periods - 1) * freq
breaks = np.linspace(start, end, periods)
if all(is_integer(x) for x in com.not_none(start, end, freq)):
# np.linspace always produces float output
breaks = maybe_downcast_to_dtype(breaks, "int64")
else:
# delegate to the appropriate range function
if isinstance(endpoint, Timestamp):
range_func = date_range
else:
range_func = timedelta_range
breaks = range_func(start=start, end=end, periods=periods, freq=freq)
return IntervalIndex.from_breaks(breaks, name=name, closed=closed)
def to_numeric(arg, errors='raise', downcast=None):
"""
Convert argument to a numeric type.
The default return dtype is `float64` or `int64`
depending on the data supplied. Use the `downcast` parameter
to obtain other dtypes.
Please note that precision loss may occur if really large numbers
are passed in. Due to the internal limitations of `ndarray`, if
numbers smaller than `-9223372036854775808` (np.iinfo(np.int64).min)
or larger than `18446744073709551615` (np.iinfo(np.uint64).max) are
passed in, it is very likely they will be converted to float so that
they can stored in an `ndarray`. These warnings apply similarly to
`Series` since it internally leverages `ndarray`.
Parameters
----------
arg : scalar, list, tuple, 1-d array, or Series
errors : {'ignore', 'raise', 'coerce'}, default 'raise'
- If 'raise', then invalid parsing will raise an exception
- If 'coerce', then invalid parsing will be set as NaN
- If 'ignore', then invalid parsing will return the input
downcast : {'integer', 'signed', 'unsigned', 'float'} , default None
If not None, and if the data has been successfully cast to a
numerical dtype (or if the data was numeric to begin with),
downcast that resulting data to the smallest numerical dtype
possible according to the following rules:
- 'integer' or 'signed': smallest signed int dtype (min.: np.int8)
- 'unsigned': smallest unsigned int dtype (min.: np.uint8)
- 'float': smallest float dtype (min.: np.float32)
As this behaviour is separate from the core conversion to
numeric values, any errors raised during the downcasting
will be surfaced regardless of the value of the 'errors' input.
In addition, downcasting will only occur if the size
of the resulting data's dtype is strictly larger than
the dtype it is to be cast to, so if none of the dtypes
checked satisfy that specification, no downcasting will be
performed on the data.
.. versionadded:: 0.19.0
Returns
-------
ret : numeric if parsing succeeded.
Return type depends on input. Series if Series, otherwise ndarray.
See Also
--------
DataFrame.astype : Cast argument to a specified dtype.
to_datetime : Convert argument to datetime.
to_timedelta : Convert argument to timedelta.
numpy.ndarray.astype : Cast a numpy array to a specified type.
Examples
--------
Take separate series and convert to numeric, coercing when told to
>>> s = pd.Series(['1.0', '2', -3])
>>> pd.to_numeric(s)
0 1.0
1 2.0
2 -3.0
dtype: float64
>>> pd.to_numeric(s, downcast='float')
0 1.0
1 2.0
2 -3.0
dtype: float32
>>> pd.to_numeric(s, downcast='signed')
0 1
1 2
2 -3
dtype: int8
>>> s = pd.Series(['apple', '1.0', '2', -3])
>>> pd.to_numeric(s, errors='ignore')
0 apple
1 1.0
2 2
3 -3
dtype: object
>>> pd.to_numeric(s, errors='coerce')
0 NaN
1 1.0
2 2.0
3 -3.0
dtype: float64
"""
if downcast not in (None, 'integer', 'signed', 'unsigned', 'float'):
raise ValueError('invalid downcasting method provided')
is_series = False
is_index = False
is_scalars = False
if isinstance(arg, ABCSeries):
is_series = True
values = arg.values
elif isinstance(arg, ABCIndexClass):
is_index = True
values = arg.asi8
if values is None:
values = arg.values
elif isinstance(arg, (list, tuple)):
values = np.array(arg, dtype='O')
elif is_scalar(arg):
if is_decimal(arg):
return float(arg)
if is_number(arg):
return arg
is_scalars = True
values = np.array([arg], dtype='O')
elif getattr(arg, 'ndim', 1) > 1:
raise TypeError('arg must be a list, tuple, 1-d array, or Series')
else:
values = arg
try:
if is_numeric_dtype(values):
pass
elif is_datetime_or_timedelta_dtype(values):
values = values.astype(np.int64)
else:
values = ensure_object(values)
coerce_numeric = errors not in ('ignore', 'raise')
values = lib.maybe_convert_numeric(values,
set(),
coerce_numeric=coerce_numeric)
except Exception:
if errors == 'raise':
raise
# attempt downcast only if the data has been successfully converted
# to a numerical dtype and if a downcast method has been specified
if downcast is not None and is_numeric_dtype(values):
typecodes = None
if downcast in ('integer', 'signed'):
typecodes = np.typecodes['Integer']
elif downcast == 'unsigned' and np.min(values) >= 0:
typecodes = np.typecodes['UnsignedInteger']
elif downcast == 'float':
typecodes = np.typecodes['Float']
# pandas support goes only to np.float32,
# as float dtypes smaller than that are
# extremely rare and not well supported
float_32_char = np.dtype(np.float32).char
float_32_ind = typecodes.index(float_32_char)
typecodes = typecodes[float_32_ind:]
if typecodes is not None:
# from smallest to largest
for dtype in typecodes:
if np.dtype(dtype).itemsize <= values.dtype.itemsize:
values = maybe_downcast_to_dtype(values, dtype)
# successful conversion
if values.dtype == dtype:
break
if is_series:
return pd.Series(values, index=arg.index, name=arg.name)
elif is_index:
# because we want to coerce to numeric if possible,
# do not use _shallow_copy_with_infer
return pd.Index(values, name=arg.name)
elif is_scalars:
return values[0]
else:
return values
def test_is_number(self):
assert is_number(True)
assert is_number(1)
assert is_number(1.1)
assert is_number(1 + 3j)
assert is_number(np.bool(False))
assert is_number(np.int64(1))
assert is_number(np.float64(1.1))
assert is_number(np.complex128(1 + 3j))
assert is_number(np.nan)
assert not is_number(None)
assert not is_number("x")
assert not is_number(datetime(2011, 1, 1))
assert not is_number(np.datetime64("2011-01-01"))
assert not is_number(Timestamp("2011-01-01"))
assert not is_number(Timestamp("2011-01-01", tz="US/Eastern"))
assert not is_number(timedelta(1000))
assert not is_number(Timedelta("1 days"))
# questionable
assert not is_number(np.bool_(False))
assert is_number(np.timedelta64(1, "D"))
def to_numeric(arg, errors='raise', downcast=None):
"""
Convert argument to a numeric type.
The default return dtype is `float64` or `int64`
depending on the data supplied. Use the `downcast` parameter
to obtain other dtypes.
Parameters
----------
arg : list, tuple, 1-d array, or Series
errors : {'ignore', 'raise', 'coerce'}, default 'raise'
- If 'raise', then invalid parsing will raise an exception
- If 'coerce', then invalid parsing will be set as NaN
- If 'ignore', then invalid parsing will return the input
downcast : {'integer', 'signed', 'unsigned', 'float'} , default None
If not None, and if the data has been successfully cast to a
numerical dtype (or if the data was numeric to begin with),
downcast that resulting data to the smallest numerical dtype
possible according to the following rules:
- 'integer' or 'signed': smallest signed int dtype (min.: np.int8)
- 'unsigned': smallest unsigned int dtype (min.: np.uint8)
- 'float': smallest float dtype (min.: np.float32)
As this behaviour is separate from the core conversion to
numeric values, any errors raised during the downcasting
will be surfaced regardless of the value of the 'errors' input.
In addition, downcasting will only occur if the size
of the resulting data's dtype is strictly larger than
the dtype it is to be cast to, so if none of the dtypes
checked satisfy that specification, no downcasting will be
performed on the data.
.. versionadded:: 0.19.0
Returns
-------
ret : numeric if parsing succeeded.
Return type depends on input. Series if Series, otherwise ndarray
Examples
--------
Take separate series and convert to numeric, coercing when told to
>>> s = pd.Series(['1.0', '2', -3])
>>> pd.to_numeric(s)
0 1.0
1 2.0
2 -3.0
dtype: float64
>>> pd.to_numeric(s, downcast='float')
0 1.0
1 2.0
2 -3.0
dtype: float32
>>> pd.to_numeric(s, downcast='signed')
0 1
1 2
2 -3
dtype: int8
>>> s = pd.Series(['apple', '1.0', '2', -3])
>>> pd.to_numeric(s, errors='ignore')
0 apple
1 1.0
2 2
3 -3
dtype: object
>>> pd.to_numeric(s, errors='coerce')
0 NaN
1 1.0
2 2.0
3 -3.0
dtype: float64
See Also
--------
pandas.DataFrame.astype : Cast argument to a specified dtype.
pandas.to_datetime : Convert argument to datetime.
pandas.to_timedelta : Convert argument to timedelta.
numpy.ndarray.astype : Cast a numpy array to a specified type.
"""
if downcast not in (None, 'integer', 'signed', 'unsigned', 'float'):
raise ValueError('invalid downcasting method provided')
is_series = False
is_index = False
is_scalars = False
if isinstance(arg, ABCSeries):
is_series = True
values = arg.values
elif isinstance(arg, ABCIndexClass):
is_index = True
values = arg.asi8
if values is None:
values = arg.values
elif isinstance(arg, (list, tuple)):
values = np.array(arg, dtype='O')
elif is_scalar(arg):
if is_decimal(arg):
return float(arg)
if is_number(arg):
return arg
is_scalars = True
values = np.array([arg], dtype='O')
elif getattr(arg, 'ndim', 1) > 1:
raise TypeError('arg must be a list, tuple, 1-d array, or Series')
else:
values = arg
try:
if is_numeric_dtype(values):
pass
elif is_datetime_or_timedelta_dtype(values):
values = values.astype(np.int64)
else:
values = ensure_object(values)
coerce_numeric = False if errors in ('ignore', 'raise') else True
values = lib.maybe_convert_numeric(values, set(),
coerce_numeric=coerce_numeric)
except Exception:
if errors == 'raise':
raise
# attempt downcast only if the data has been successfully converted
# to a numerical dtype and if a downcast method has been specified
if downcast is not None and is_numeric_dtype(values):
typecodes = None
if downcast in ('integer', 'signed'):
typecodes = np.typecodes['Integer']
elif downcast == 'unsigned' and np.min(values) >= 0:
typecodes = np.typecodes['UnsignedInteger']
elif downcast == 'float':
typecodes = np.typecodes['Float']
# pandas support goes only to np.float32,
# as float dtypes smaller than that are
# extremely rare and not well supported
float_32_char = np.dtype(np.float32).char
float_32_ind = typecodes.index(float_32_char)
typecodes = typecodes[float_32_ind:]
if typecodes is not None:
# from smallest to largest
for dtype in typecodes:
if np.dtype(dtype).itemsize <= values.dtype.itemsize:
values = maybe_downcast_to_dtype(values, dtype)
# successful conversion
if values.dtype == dtype:
break
if is_series:
return pd.Series(values, index=arg.index, name=arg.name)
elif is_index:
# because we want to coerce to numeric if possible,
# do not use _shallow_copy_with_infer
return pd.Index(values, name=arg.name)
elif is_scalars:
return values[0]
else:
return values
def interval_range(start=None, end=None, periods=None, freq=None,
name=None, closed='right'):
"""
Return a fixed frequency IntervalIndex
Parameters
----------
start : numeric or datetime-like, default None
Left bound for generating intervals
end : numeric or datetime-like, default None
Right bound for generating intervals
periods : integer, default None
Number of periods to generate
freq : numeric, string, or DateOffset, default None
The length of each interval. Must be consistent with the type of start
and end, e.g. 2 for numeric, or '5H' for datetime-like. Default is 1
for numeric and 'D' for datetime-like.
name : string, default None
Name of the resulting IntervalIndex
closed : {'left', 'right', 'both', 'neither'}, default 'right'
Whether the intervals are closed on the left-side, right-side, both
or neither.
Returns
-------
rng : IntervalIndex
See Also
--------
IntervalIndex : An Index of intervals that are all closed on the same side.
Notes
-----
Of the four parameters ``start``, ``end``, ``periods``, and ``freq``,
exactly three must be specified. If ``freq`` is omitted, the resulting
``IntervalIndex`` will have ``periods`` linearly spaced elements between
``start`` and ``end``, inclusively.
To learn more about datetime-like frequency strings, please see `this link
<http://pandas.pydata.org/pandas-docs/stable/timeseries.html#offset-aliases>`__.
Examples
--------
Numeric ``start`` and ``end`` is supported.
>>> pd.interval_range(start=0, end=5)
IntervalIndex([(0, 1], (1, 2], (2, 3], (3, 4], (4, 5]],
closed='right', dtype='interval[int64]')
Additionally, datetime-like input is also supported.
>>> pd.interval_range(start=pd.Timestamp('2017-01-01'),
... end=pd.Timestamp('2017-01-04'))
IntervalIndex([(2017-01-01, 2017-01-02], (2017-01-02, 2017-01-03],
(2017-01-03, 2017-01-04]],
closed='right', dtype='interval[datetime64[ns]]')
The ``freq`` parameter specifies the frequency between the left and right.
endpoints of the individual intervals within the ``IntervalIndex``. For
numeric ``start`` and ``end``, the frequency must also be numeric.
>>> pd.interval_range(start=0, periods=4, freq=1.5)
IntervalIndex([(0.0, 1.5], (1.5, 3.0], (3.0, 4.5], (4.5, 6.0]],
closed='right', dtype='interval[float64]')
Similarly, for datetime-like ``start`` and ``end``, the frequency must be
convertible to a DateOffset.
>>> pd.interval_range(start=pd.Timestamp('2017-01-01'),
... periods=3, freq='MS')
IntervalIndex([(2017-01-01, 2017-02-01], (2017-02-01, 2017-03-01],
(2017-03-01, 2017-04-01]],
closed='right', dtype='interval[datetime64[ns]]')
Specify ``start``, ``end``, and ``periods``; the frequency is generated
automatically (linearly spaced).
>>> pd.interval_range(start=0, end=6, periods=4)
IntervalIndex([(0.0, 1.5], (1.5, 3.0], (3.0, 4.5], (4.5, 6.0]],
closed='right',
dtype='interval[float64]')
The ``closed`` parameter specifies which endpoints of the individual
intervals within the ``IntervalIndex`` are closed.
>>> pd.interval_range(end=5, periods=4, closed='both')
IntervalIndex([[1, 2], [2, 3], [3, 4], [4, 5]],
closed='both', dtype='interval[int64]')
"""
start = com.maybe_box_datetimelike(start)
end = com.maybe_box_datetimelike(end)
endpoint = start if start is not None else end
if freq is None and com._any_none(periods, start, end):
freq = 1 if is_number(endpoint) else 'D'
if com.count_not_none(start, end, periods, freq) != 3:
raise ValueError('Of the four parameters: start, end, periods, and '
'freq, exactly three must be specified')
if not _is_valid_endpoint(start):
msg = 'start must be numeric or datetime-like, got {start}'
raise ValueError(msg.format(start=start))
elif not _is_valid_endpoint(end):
msg = 'end must be numeric or datetime-like, got {end}'
raise ValueError(msg.format(end=end))
if is_float(periods):
periods = int(periods)
elif not is_integer(periods) and periods is not None:
msg = 'periods must be a number, got {periods}'
raise TypeError(msg.format(periods=periods))
if freq is not None and not is_number(freq):
try:
freq = to_offset(freq)
except ValueError:
raise ValueError('freq must be numeric or convertible to '
'DateOffset, got {freq}'.format(freq=freq))
# verify type compatibility
if not all([_is_type_compatible(start, end),
_is_type_compatible(start, freq),
_is_type_compatible(end, freq)]):
raise TypeError("start, end, freq need to be type compatible")
# +1 to convert interval count to breaks count (n breaks = n-1 intervals)
if periods is not None:
periods += 1
if is_number(endpoint):
# force consistency between start/end/freq (lower end if freq skips it)
if com._all_not_none(start, end, freq):
end -= (end - start) % freq
# compute the period/start/end if unspecified (at most one)
if periods is None:
periods = int((end - start) // freq) + 1
elif start is None:
start = end - (periods - 1) * freq
elif end is None:
end = start + (periods - 1) * freq
breaks = np.linspace(start, end, periods)
if all(is_integer(x) for x in com._not_none(start, end, freq)):
# np.linspace always produces float output
breaks = maybe_downcast_to_dtype(breaks, 'int64')
else:
# delegate to the appropriate range function
if isinstance(endpoint, Timestamp):
range_func = date_range
else:
range_func = timedelta_range
breaks = range_func(start=start, end=end, periods=periods, freq=freq)
return IntervalIndex.from_breaks(breaks, name=name, closed=closed)
def _is_numeric(x):
""" Whether the array x is numeric. """
return all(is_number(i) for i in x)
def assert_almost_equal(
left,
right,
check_dtype: bool | str = "equiv",
check_less_precise: bool | int | NoDefault = no_default,
rtol: float = 1.0e-5,
atol: float = 1.0e-8,
**kwargs,
):
"""
Check that the left and right objects are approximately equal.
By approximately equal, we refer to objects that are numbers or that
contain numbers which may be equivalent to specific levels of precision.
Parameters
----------
left : object
right : object
check_dtype : bool or {'equiv'}, default 'equiv'
Check dtype if both a and b are the same type. If 'equiv' is passed in,
then `RangeIndex` and `Int64Index` are also considered equivalent
when doing type checking.
check_less_precise : bool or int, default False
Specify comparison precision. 5 digits (False) or 3 digits (True)
after decimal points are compared. If int, then specify the number
of digits to compare.
When comparing two numbers, if the first number has magnitude less
than 1e-5, we compare the two numbers directly and check whether
they are equivalent within the specified precision. Otherwise, we
compare the **ratio** of the second number to the first number and
check whether it is equivalent to 1 within the specified precision.
.. deprecated:: 1.1.0
Use `rtol` and `atol` instead to define relative/absolute
tolerance, respectively. Similar to :func:`math.isclose`.
rtol : float, default 1e-5
Relative tolerance.
.. versionadded:: 1.1.0
atol : float, default 1e-8
Absolute tolerance.
.. versionadded:: 1.1.0
"""
if check_less_precise is not no_default:
warnings.warn(
"The 'check_less_precise' keyword in testing.assert_*_equal "
"is deprecated and will be removed in a future version. "
"You can stop passing 'check_less_precise' to silence this warning.",
FutureWarning,
stacklevel=find_stack_level(),
)
# https://github.com/python/mypy/issues/7642
# error: Argument 1 to "_get_tol_from_less_precise" has incompatible
# type "Union[bool, int, NoDefault]"; expected "Union[bool, int]"
rtol = atol = _get_tol_from_less_precise(
check_less_precise # type: ignore[arg-type]
)
if isinstance(left, Index):
assert_index_equal(
left,
right,
check_exact=False,
exact=check_dtype,
rtol=rtol,
atol=atol,
**kwargs,
)
elif isinstance(left, Series):
assert_series_equal(
left,
right,
check_exact=False,
check_dtype=check_dtype,
rtol=rtol,
atol=atol,
**kwargs,
)
elif isinstance(left, DataFrame):
assert_frame_equal(
left,
right,
check_exact=False,
check_dtype=check_dtype,
rtol=rtol,
atol=atol,
**kwargs,
)
else:
# Other sequences.
if check_dtype:
if is_number(left) and is_number(right):
# Do not compare numeric classes, like np.float64 and float.
pass
elif is_bool(left) and is_bool(right):
# Do not compare bool classes, like np.bool_ and bool.
pass
else:
if isinstance(left, np.ndarray) or isinstance(
right, np.ndarray):
obj = "numpy array"
else:
obj = "Input"
assert_class_equal(left, right, obj=obj)
# if we have "equiv", this becomes True
check_dtype = bool(check_dtype)
_testing.assert_almost_equal(left,
right,
check_dtype=check_dtype,
rtol=rtol,
atol=atol,
**kwargs)
def to_numeric(arg, errors="raise", downcast=None):
"""
Convert argument to a numeric type.
The default return dtype is `float64` or `int64`
depending on the data supplied. Use the `downcast` parameter
to obtain other dtypes.
Please note that precision loss may occur if really large numbers
are passed in. Due to the internal limitations of `ndarray`, if
numbers smaller than `-9223372036854775808` (np.iinfo(np.int64).min)
or larger than `18446744073709551615` (np.iinfo(np.uint64).max) are
passed in, it is very likely they will be converted to float so that
they can stored in an `ndarray`. These warnings apply similarly to
`Series` since it internally leverages `ndarray`.
Parameters
----------
arg : scalar, list, tuple, 1-d array, or Series
Argument to be converted.
errors : {'ignore', 'raise', 'coerce'}, default 'raise'
- If 'raise', then invalid parsing will raise an exception.
- If 'coerce', then invalid parsing will be set as NaN.
- If 'ignore', then invalid parsing will return the input.
downcast : {'integer', 'signed', 'unsigned', 'float'}, default None
If not None, and if the data has been successfully cast to a
numerical dtype (or if the data was numeric to begin with),
downcast that resulting data to the smallest numerical dtype
possible according to the following rules:
- 'integer' or 'signed': smallest signed int dtype (min.: np.int8)
- 'unsigned': smallest unsigned int dtype (min.: np.uint8)
- 'float': smallest float dtype (min.: np.float32)
As this behaviour is separate from the core conversion to
numeric values, any errors raised during the downcasting
will be surfaced regardless of the value of the 'errors' input.
In addition, downcasting will only occur if the size
of the resulting data's dtype is strictly larger than
the dtype it is to be cast to, so if none of the dtypes
checked satisfy that specification, no downcasting will be
performed on the data.
Returns
-------
ret
Numeric if parsing succeeded.
Return type depends on input. Series if Series, otherwise ndarray.
See Also
--------
DataFrame.astype : Cast argument to a specified dtype.
to_datetime : Convert argument to datetime.
to_timedelta : Convert argument to timedelta.
numpy.ndarray.astype : Cast a numpy array to a specified type.
DataFrame.convert_dtypes : Convert dtypes.
Examples
--------
Take separate series and convert to numeric, coercing when told to
>>> s = pd.Series(['1.0', '2', -3])
>>> pd.to_numeric(s)
0 1.0
1 2.0
2 -3.0
dtype: float64
>>> pd.to_numeric(s, downcast='float')
0 1.0
1 2.0
2 -3.0
dtype: float32
>>> pd.to_numeric(s, downcast='signed')
0 1
1 2
2 -3
dtype: int8
>>> s = pd.Series(['apple', '1.0', '2', -3])
>>> pd.to_numeric(s, errors='ignore')
0 apple
1 1.0
2 2
3 -3
dtype: object
>>> pd.to_numeric(s, errors='coerce')
0 NaN
1 1.0
2 2.0
3 -3.0
dtype: float64
Downcasting of nullable integer and floating dtypes is supported:
>>> s = pd.Series([1, 2, 3], dtype="Int64")
>>> pd.to_numeric(s, downcast="integer")
0 1
1 2
2 3
dtype: Int8
>>> s = pd.Series([1.0, 2.1, 3.0], dtype="Float64")
>>> pd.to_numeric(s, downcast="float")
0 1.0
1 2.1
2 3.0
dtype: Float32
"""
if downcast not in (None, "integer", "signed", "unsigned", "float"):
raise ValueError("invalid downcasting method provided")
if errors not in ("ignore", "raise", "coerce"):
raise ValueError("invalid error value specified")
is_series = False
is_index = False
is_scalars = False
if isinstance(arg, ABCSeries):
is_series = True
values = arg.values
elif isinstance(arg, ABCIndex):
is_index = True
if needs_i8_conversion(arg.dtype):
values = arg.asi8
else:
values = arg.values
elif isinstance(arg, (list, tuple)):
values = np.array(arg, dtype="O")
elif is_scalar(arg):
if is_decimal(arg):
return float(arg)
if is_number(arg):
return arg
is_scalars = True
values = np.array([arg], dtype="O")
elif getattr(arg, "ndim", 1) > 1:
raise TypeError("arg must be a list, tuple, 1-d array, or Series")
else:
values = arg
# GH33013: for IntegerArray & FloatingArray extract non-null values for casting
# save mask to reconstruct the full array after casting
if isinstance(values, NumericArray):
mask = values._mask
values = values._data[~mask]
else:
mask = None
values_dtype = getattr(values, "dtype", None)
if is_numeric_dtype(values_dtype):
pass
elif is_datetime_or_timedelta_dtype(values_dtype):
values = values.view(np.int64)
else:
values = ensure_object(values)
coerce_numeric = errors not in ("ignore", "raise")
try:
values = lib.maybe_convert_numeric(values,
set(),
coerce_numeric=coerce_numeric)
except (ValueError, TypeError):
if errors == "raise":
raise
# attempt downcast only if the data has been successfully converted
# to a numerical dtype and if a downcast method has been specified
if downcast is not None and is_numeric_dtype(values.dtype):
typecodes = None
if downcast in ("integer", "signed"):
typecodes = np.typecodes["Integer"]
elif downcast == "unsigned" and (not len(values)
or np.min(values) >= 0):
typecodes = np.typecodes["UnsignedInteger"]
elif downcast == "float":
typecodes = np.typecodes["Float"]
# pandas support goes only to np.float32,
# as float dtypes smaller than that are
# extremely rare and not well supported
float_32_char = np.dtype(np.float32).char
float_32_ind = typecodes.index(float_32_char)
typecodes = typecodes[float_32_ind:]
if typecodes is not None:
# from smallest to largest
for dtype in typecodes:
dtype = np.dtype(dtype)
if dtype.itemsize <= values.dtype.itemsize:
values = maybe_downcast_numeric(values, dtype)
# successful conversion
if values.dtype == dtype:
break
# GH33013: for IntegerArray & FloatingArray need to reconstruct masked array
if mask is not None:
data = np.zeros(mask.shape, dtype=values.dtype)
data[~mask] = values
from pandas.core.arrays import FloatingArray, IntegerArray
klass = IntegerArray if is_integer_dtype(data.dtype) else FloatingArray
values = klass(data, mask)
if is_series:
return arg._constructor(values, index=arg.index, name=arg.name)
elif is_index:
# because we want to coerce to numeric if possible,
# do not use _shallow_copy
return pd.Index(values, name=arg.name)
elif is_scalars:
return values[0]
else:
return values
def interval_range(start=None, end=None, periods=None, freq=None,
name=None, closed='right'):
"""
Return a fixed frequency IntervalIndex
Parameters
----------
start : numeric or datetime-like, default None
Left bound for generating intervals
end : numeric or datetime-like, default None
Right bound for generating intervals
periods : integer, default None
Number of periods to generate
freq : numeric, string, or DateOffset, default None
The length of each interval. Must be consistent with the type of start
and end, e.g. 2 for numeric, or '5H' for datetime-like. Default is 1
for numeric and 'D' (calendar daily) for datetime-like.
name : string, default None
Name of the resulting IntervalIndex
closed : string, default 'right'
options are: 'left', 'right', 'both', 'neither'
Notes
-----
Of the three parameters: ``start``, ``end``, and ``periods``, exactly two
must be specified.
Returns
-------
rng : IntervalIndex
Examples
--------
Numeric ``start`` and ``end`` is supported.
>>> pd.interval_range(start=0, end=5)
IntervalIndex([(0, 1], (1, 2], (2, 3], (3, 4], (4, 5]]
closed='right', dtype='interval[int64]')
Additionally, datetime-like input is also supported.
>>> pd.interval_range(start=pd.Timestamp('2017-01-01'),
end=pd.Timestamp('2017-01-04'))
IntervalIndex([(2017-01-01, 2017-01-02], (2017-01-02, 2017-01-03],
(2017-01-03, 2017-01-04]]
closed='right', dtype='interval[datetime64[ns]]')
The ``freq`` parameter specifies the frequency between the left and right.
endpoints of the individual intervals within the ``IntervalIndex``. For
numeric ``start`` and ``end``, the frequency must also be numeric.
>>> pd.interval_range(start=0, periods=4, freq=1.5)
IntervalIndex([(0.0, 1.5], (1.5, 3.0], (3.0, 4.5], (4.5, 6.0]]
closed='right', dtype='interval[float64]')
Similarly, for datetime-like ``start`` and ``end``, the frequency must be
convertible to a DateOffset.
>>> pd.interval_range(start=pd.Timestamp('2017-01-01'),
periods=3, freq='MS')
IntervalIndex([(2017-01-01, 2017-02-01], (2017-02-01, 2017-03-01],
(2017-03-01, 2017-04-01]]
closed='right', dtype='interval[datetime64[ns]]')
The ``closed`` parameter specifies which endpoints of the individual
intervals within the ``IntervalIndex`` are closed.
>>> pd.interval_range(end=5, periods=4, closed='both')
IntervalIndex([[1, 2], [2, 3], [3, 4], [4, 5]]
closed='both', dtype='interval[int64]')
See Also
--------
IntervalIndex : an Index of intervals that are all closed on the same side.
"""
if _count_not_none(start, end, periods) != 2:
raise ValueError('Of the three parameters: start, end, and periods, '
'exactly two must be specified')
start = _maybe_box_datetimelike(start)
end = _maybe_box_datetimelike(end)
endpoint = next(_not_none(start, end))
if not _is_valid_endpoint(start):
msg = 'start must be numeric or datetime-like, got {start}'
raise ValueError(msg.format(start=start))
if not _is_valid_endpoint(end):
msg = 'end must be numeric or datetime-like, got {end}'
raise ValueError(msg.format(end=end))
if is_float(periods):
periods = int(periods)
elif not is_integer(periods) and periods is not None:
msg = 'periods must be a number, got {periods}'
raise TypeError(msg.format(periods=periods))
freq = freq or (1 if is_number(endpoint) else 'D')
if not is_number(freq):
try:
freq = to_offset(freq)
except ValueError:
raise ValueError('freq must be numeric or convertible to '
'DateOffset, got {freq}'.format(freq=freq))
# verify type compatibility
if not all([_is_type_compatible(start, end),
_is_type_compatible(start, freq),
_is_type_compatible(end, freq)]):
raise TypeError("start, end, freq need to be type compatible")
if is_number(endpoint):
if periods is None:
periods = int((end - start) // freq)
if start is None:
start = end - periods * freq
# force end to be consistent with freq (lower if freq skips over end)
end = start + periods * freq
# end + freq for inclusive endpoint
breaks = np.arange(start, end + freq, freq)
elif isinstance(endpoint, Timestamp):
# add one to account for interval endpoints (n breaks = n-1 intervals)
if periods is not None:
periods += 1
breaks = date_range(start=start, end=end, periods=periods, freq=freq)
else:
# add one to account for interval endpoints (n breaks = n-1 intervals)
if periods is not None:
periods += 1
breaks = timedelta_range(start=start, end=end, periods=periods,
freq=freq)
return IntervalIndex.from_breaks(breaks, name=name, closed=closed)
def test_is_number(self):
assert is_number(True)
assert is_number(1)
assert is_number(1.1)
assert is_number(1 + 3j)
assert is_number(np.bool(False))
assert is_number(np.int64(1))
assert is_number(np.float64(1.1))
assert is_number(np.complex128(1 + 3j))
assert is_number(np.nan)
assert not is_number(None)
assert not is_number('x')
assert not is_number(datetime(2011, 1, 1))
assert not is_number(np.datetime64('2011-01-01'))
assert not is_number(Timestamp('2011-01-01'))
assert not is_number(Timestamp('2011-01-01', tz='US/Eastern'))
assert not is_number(timedelta(1000))
assert not is_number(Timedelta('1 days'))
# questionable
assert not is_number(np.bool_(False))
assert is_number(np.timedelta64(1, 'D'))
def _parse_errorbars(self, label, err):
"""
Look for error keyword arguments and return the actual errorbar data
or return the error DataFrame/dict
Error bars can be specified in several ways:
Series: the user provides a pandas.Series object of the same
length as the data
ndarray: provides a np.ndarray of the same length as the data
DataFrame/dict: error values are paired with keys matching the
key in the plotted DataFrame
str: the name of the column within the plotted DataFrame
"""
if err is None:
return None
def match_labels(data, e):
e = e.reindex(data.index)
return e
# key-matched DataFrame
if isinstance(err, ABCDataFrame):
err = match_labels(self.data, err)
# key-matched dict
elif isinstance(err, dict):
pass
# Series of error values
elif isinstance(err, ABCSeries):
# broadcast error series across data
err = match_labels(self.data, err)
err = np.atleast_2d(err)
err = np.tile(err, (self.nseries, 1))
# errors are a column in the dataframe
elif isinstance(err, str):
evalues = self.data[err].values
self.data = self.data[self.data.columns.drop(err)]
err = np.atleast_2d(evalues)
err = np.tile(err, (self.nseries, 1))
elif is_list_like(err):
if is_iterator(err):
err = np.atleast_2d(list(err))
else:
# raw error values
err = np.atleast_2d(err)
err_shape = err.shape
# asymmetrical error bars
if err.ndim == 3:
if ((err_shape[0] != self.nseries) or (err_shape[1] != 2)
or (err_shape[2] != len(self.data))):
msg = ("Asymmetrical error bars should be provided " +
"with the shape (%u, 2, %u)" %
(self.nseries, len(self.data)))
raise ValueError(msg)
# broadcast errors to each data series
if len(err) == 1:
err = np.tile(err, (self.nseries, 1))
elif is_number(err):
err = np.tile([err], (self.nseries, len(self.data)))
else:
msg = "No valid {label} detected".format(label=label)
raise ValueError(msg)
return err
def _is_valid_endpoint(endpoint):
"""helper for interval_range to check if start/end are valid types"""
return any([is_number(endpoint),
isinstance(endpoint, Timestamp),
isinstance(endpoint, Timedelta),
endpoint is None])
