def __init__(self,
date_column,
step_size=None,
train_size=None,
test_size=None):
if not isinstance(step_size, timedelta):
raise ValueError('step_size should be of type datetime.timedelta')
if not isinstance(train_size, timedelta):
raise ValueError('train_size should be of type datetime.timedelta')
if not isinstance(test_size, timedelta):
raise ValueError('test_size should be of type datetime.timedelta')
if not isinstance(date_column, DatetimeArray):
try:
date_column = DatetimeArray(datecolumn)
except:
print(
"The provided date column cannot be parsed. Please provide a datetime column in pandas DatetimeArray format."
)
self.date_column = date_column
self.step_size = step_size
self.train_size = train_size
self.test_size = test_size
self.fold_size = self.train_size + self.test_size
def _to_datetime_with_format(
arg,
orig_arg,
name,
tz,
fmt: str,
exact: bool,
errors: Optional[str],
infer_datetime_format: bool,
) -> Optional[Index]:
"""
Try parsing with the given format, returning None on failure.
"""
result = None
try:
# shortcut formatting here
if fmt == "%Y%m%d":
# pass orig_arg as float-dtype may have been converted to
# datetime64[ns]
orig_arg = ensure_object(orig_arg)
try:
# may return None without raising
result = _attempt_YYYYMMDD(orig_arg, errors=errors)
except (ValueError, TypeError, OutOfBoundsDatetime) as err:
raise ValueError(
"cannot convert the input to '%Y%m%d' date format"
) from err
if result is not None:
utc = tz == "utc"
return _box_as_indexlike(result, utc=utc, name=name)
# fallback
if result is None:
# error: Incompatible types in assignment (expression has type
# "Optional[Index]", variable has type "Optional[ndarray]")
result = _array_strptime_with_fallback( # type: ignore[assignment]
arg, name, tz, fmt, exact, errors, infer_datetime_format
)
if result is not None:
return result
except ValueError as e:
# Fallback to try to convert datetime objects if timezone-aware
# datetime objects are found without passing `utc=True`
try:
values, tz = conversion.datetime_to_datetime64(arg)
dta = DatetimeArray(values, dtype=tz_to_dtype(tz))
return DatetimeIndex._simple_new(dta, name=name)
except (ValueError, TypeError):
raise e
# error: Incompatible return value type (got "Optional[ndarray]", expected
# "Optional[Index]")
return result # type: ignore[return-value]
def _to_datetime_with_format(
arg,
orig_arg,
name,
tz,
fmt: str,
exact: bool,
errors: str,
infer_datetime_format: bool,
) -> Index | None:
"""
Try parsing with the given format, returning None on failure.
"""
result = None
try:
# shortcut formatting here
if fmt == "%Y%m%d":
# pass orig_arg as float-dtype may have been converted to
# datetime64[ns]
orig_arg = ensure_object(orig_arg)
try:
# may return None without raising
result = _attempt_YYYYMMDD(orig_arg, errors=errors)
except (ValueError, TypeError, OutOfBoundsDatetime) as err:
raise ValueError(
"cannot convert the input to '%Y%m%d' date format"
) from err
if result is not None:
utc = tz == "utc"
return _box_as_indexlike(result, utc=utc, name=name)
# fallback
res = _array_strptime_with_fallback(
arg, name, tz, fmt, exact, errors, infer_datetime_format
)
return res
except ValueError as err:
# Fallback to try to convert datetime objects if timezone-aware
# datetime objects are found without passing `utc=True`
try:
values, tz = conversion.datetime_to_datetime64(arg)
dta = DatetimeArray(values, dtype=tz_to_dtype(tz))
return DatetimeIndex._simple_new(dta, name=name)
except (ValueError, TypeError):
raise err
def array(self) -> ExtensionArray:
"""
The ExtensionArray of the data backing this Series or Index.
.. versionadded:: 0.24.0
Returns
-------
ExtensionArray
An ExtensionArray of the values stored within. For extension
types, this is the actual array. For NumPy native types, this
is a thin (no copy) wrapper around :class:`numpy.ndarray`.
``.array`` differs ``.values`` which may require converting the
data to a different form.
See Also
--------
Index.to_numpy : Similar method that always returns a NumPy array.
Series.to_numpy : Similar method that always returns a NumPy array.
Notes
-----
This table lays out the different array types for each extension
dtype within pandas.
================== =============================
dtype array type
================== =============================
category Categorical
period PeriodArray
interval IntervalArray
IntegerNA IntegerArray
datetime64[ns, tz] DatetimeArray
================== =============================
For any 3rd-party extension types, the array type will be an
ExtensionArray.
For all remaining dtypes ``.array`` will be a
:class:`arrays.NumpyExtensionArray` wrapping the actual ndarray
stored within. If you absolutely need a NumPy array (possibly with
copying / coercing data), then use :meth:`Series.to_numpy` instead.
Examples
--------
For regular NumPy types like int, and float, a PandasArray
is returned.
>>> pd.Series([1, 2, 3]).array
<PandasArray>
[1, 2, 3]
Length: 3, dtype: int64
For extension types, like Categorical, the actual ExtensionArray
is returned
>>> ser = pd.Series(pd.Categorical(['a', 'b', 'a']))
>>> ser.array
[a, b, a]
Categories (2, object): [a, b]
"""
# As a mixin, we depend on the mixing class having _values.
# Special mixin syntax may be developed in the future:
# https://github.com/python/typing/issues/246
result = self._values # type: ignore
if is_datetime64_ns_dtype(result.dtype):
from pandas.arrays import DatetimeArray
result = DatetimeArray(result)
elif is_timedelta64_ns_dtype(result.dtype):
from pandas.arrays import TimedeltaArray
result = TimedeltaArray(result)
elif not is_extension_array_dtype(result.dtype):
from pandas.core.arrays.numpy_ import PandasArray
result = PandasArray(result)
return result
def _convert_listlike_datetimes(
arg,
format: str | None,
name: Hashable = None,
tz: Timezone | None = None,
unit: str | None = None,
errors: str = "raise",
infer_datetime_format: bool = False,
dayfirst: bool | None = None,
yearfirst: bool | None = None,
exact: bool = True,
):
"""
Helper function for to_datetime. Performs the conversions of 1D listlike
of dates
Parameters
----------
arg : list, tuple, ndarray, Series, Index
date to be parsed
name : object
None or string for the Index name
tz : object
None or 'utc'
unit : str
None or string of the frequency of the passed data
errors : str
error handing behaviors from to_datetime, 'raise', 'coerce', 'ignore'
infer_datetime_format : bool, default False
inferring format behavior from to_datetime
dayfirst : bool
dayfirst parsing behavior from to_datetime
yearfirst : bool
yearfirst parsing behavior from to_datetime
exact : bool, default True
exact format matching behavior from to_datetime
Returns
-------
Index-like of parsed dates
"""
if isinstance(arg, (list, tuple)):
arg = np.array(arg, dtype="O")
arg_dtype = getattr(arg, "dtype", None)
# these are shortcutable
if is_datetime64tz_dtype(arg_dtype):
if not isinstance(arg, (DatetimeArray, DatetimeIndex)):
return DatetimeIndex(arg, tz=tz, name=name)
if tz == "utc":
arg = arg.tz_convert(None).tz_localize(tz)
return arg
elif is_datetime64_ns_dtype(arg_dtype):
if not isinstance(arg, (DatetimeArray, DatetimeIndex)):
try:
return DatetimeIndex(arg, tz=tz, name=name)
except ValueError:
pass
elif tz:
# DatetimeArray, DatetimeIndex
return arg.tz_localize(tz)
return arg
elif unit is not None:
if format is not None:
raise ValueError("cannot specify both format and unit")
return _to_datetime_with_unit(arg, unit, name, tz, errors)
elif getattr(arg, "ndim", 1) > 1:
raise TypeError(
"arg must be a string, datetime, list, tuple, 1-d array, or Series"
)
# warn if passing timedelta64, raise for PeriodDtype
# NB: this must come after unit transformation
orig_arg = arg
try:
arg, _ = maybe_convert_dtype(arg, copy=False)
except TypeError:
if errors == "coerce":
npvalues = np.array(["NaT"],
dtype="datetime64[ns]").repeat(len(arg))
return DatetimeIndex(npvalues, name=name)
elif errors == "ignore":
idx = Index(arg, name=name)
return idx
raise
arg = ensure_object(arg)
require_iso8601 = False
if infer_datetime_format and format is None:
format = _guess_datetime_format_for_array(arg, dayfirst=dayfirst)
if format is not None:
# There is a special fast-path for iso8601 formatted
# datetime strings, so in those cases don't use the inferred
# format because this path makes process slower in this
# special case
format_is_iso8601 = format_is_iso(format)
if format_is_iso8601:
require_iso8601 = not infer_datetime_format
format = None
if format is not None:
res = _to_datetime_with_format(arg, orig_arg, name, tz, format, exact,
errors, infer_datetime_format)
if res is not None:
return res
assert format is None or infer_datetime_format
utc = tz == "utc"
result, tz_parsed = objects_to_datetime64ns(
arg,
dayfirst=dayfirst,
yearfirst=yearfirst,
utc=utc,
errors=errors,
require_iso8601=require_iso8601,
allow_object=True,
)
if tz_parsed is not None:
# We can take a shortcut since the datetime64 numpy array
# is in UTC
dta = DatetimeArray(result, dtype=tz_to_dtype(tz_parsed))
return DatetimeIndex._simple_new(dta, name=name)
utc = tz == "utc"
return _box_as_indexlike(result, utc=utc, name=name)
def _convert_listlike_datetimes(
arg,
format,
name=None,
tz=None,
unit=None,
errors=None,
infer_datetime_format=None,
dayfirst=None,
yearfirst=None,
exact=None,
):
"""
Helper function for to_datetime. Performs the conversions of 1D listlike
of dates
Parameters
----------
arg : list, tuple, ndarray, Series, Index
date to be parsed
name : object
None or string for the Index name
tz : object
None or 'utc'
unit : string
None or string of the frequency of the passed data
errors : string
error handing behaviors from to_datetime, 'raise', 'coerce', 'ignore'
infer_datetime_format : boolean
inferring format behavior from to_datetime
dayfirst : boolean
dayfirst parsing behavior from to_datetime
yearfirst : boolean
yearfirst parsing behavior from to_datetime
exact : boolean
exact format matching behavior from to_datetime
Returns
-------
Index-like of parsed dates
"""
if isinstance(arg, (list, tuple)):
arg = np.array(arg, dtype="O")
arg_dtype = getattr(arg, "dtype", None)
# these are shortcutable
if is_datetime64tz_dtype(arg_dtype):
if not isinstance(arg, (DatetimeArray, DatetimeIndex)):
return DatetimeIndex(arg, tz=tz, name=name)
if tz == "utc":
# error: Item "DatetimeIndex" of "Union[DatetimeArray, DatetimeIndex]" has
# no attribute "tz_convert"
arg = arg.tz_convert(None).tz_localize(tz) # type: ignore
return arg
elif is_datetime64_ns_dtype(arg_dtype):
if not isinstance(arg, (DatetimeArray, DatetimeIndex)):
try:
return DatetimeIndex(arg, tz=tz, name=name)
except ValueError:
pass
elif tz:
# DatetimeArray, DatetimeIndex
# error: Item "DatetimeIndex" of "Union[DatetimeArray, DatetimeIndex]" has
# no attribute "tz_localize"
return arg.tz_localize(tz) # type: ignore
return arg
elif unit is not None:
if format is not None:
raise ValueError("cannot specify both format and unit")
arg = getattr(arg, "_values", arg)
# GH 30050 pass an ndarray to tslib.array_with_unit_to_datetime
# because it expects an ndarray argument
if isinstance(arg, IntegerArray):
result = arg.astype(f"datetime64[{unit}]")
tz_parsed = None
else:
result, tz_parsed = tslib.array_with_unit_to_datetime(
arg, unit, errors=errors)
if errors == "ignore":
result = Index(result, name=name)
else:
result = DatetimeIndex(result, name=name)
# GH 23758: We may still need to localize the result with tz
# GH 25546: Apply tz_parsed first (from arg), then tz (from caller)
# result will be naive but in UTC
try:
result = result.tz_localize("UTC").tz_convert(tz_parsed)
except AttributeError:
# Regular Index from 'ignore' path
return result
if tz is not None:
if result.tz is None:
result = result.tz_localize(tz)
else:
result = result.tz_convert(tz)
return result
elif getattr(arg, "ndim", 1) > 1:
raise TypeError(
"arg must be a string, datetime, list, tuple, 1-d array, or Series"
)
# warn if passing timedelta64, raise for PeriodDtype
# NB: this must come after unit transformation
orig_arg = arg
try:
arg, _ = maybe_convert_dtype(arg, copy=False)
except TypeError:
if errors == "coerce":
result = np.array(["NaT"], dtype="datetime64[ns]").repeat(len(arg))
return DatetimeIndex(result, name=name)
elif errors == "ignore":
result = Index(arg, name=name)
return result
raise
arg = ensure_object(arg)
require_iso8601 = False
if infer_datetime_format and format is None:
format = _guess_datetime_format_for_array(arg, dayfirst=dayfirst)
if format is not None:
# There is a special fast-path for iso8601 formatted
# datetime strings, so in those cases don't use the inferred
# format because this path makes process slower in this
# special case
format_is_iso8601 = _format_is_iso(format)
if format_is_iso8601:
require_iso8601 = not infer_datetime_format
format = None
tz_parsed = None
result = None
if format is not None:
try:
# shortcut formatting here
if format == "%Y%m%d":
try:
# pass orig_arg as float-dtype may have been converted to
# datetime64[ns]
orig_arg = ensure_object(orig_arg)
result = _attempt_YYYYMMDD(orig_arg, errors=errors)
except (ValueError, TypeError,
tslibs.OutOfBoundsDatetime) as err:
raise ValueError(
"cannot convert the input to '%Y%m%d' date format"
) from err
# fallback
if result is None:
try:
result, timezones = array_strptime(arg,
format,
exact=exact,
errors=errors)
if "%Z" in format or "%z" in format:
return _return_parsed_timezone_results(
result, timezones, tz, name)
except tslibs.OutOfBoundsDatetime:
if errors == "raise":
raise
elif errors == "coerce":
result = np.empty(arg.shape, dtype="M8[ns]")
iresult = result.view("i8")
iresult.fill(tslibs.iNaT)
else:
result = arg
except ValueError:
# if format was inferred, try falling back
# to array_to_datetime - terminate here
# for specified formats
if not infer_datetime_format:
if errors == "raise":
raise
elif errors == "coerce":
result = np.empty(arg.shape, dtype="M8[ns]")
iresult = result.view("i8")
iresult.fill(tslibs.iNaT)
else:
result = arg
except ValueError as e:
# Fallback to try to convert datetime objects if timezone-aware
# datetime objects are found without passing `utc=True`
try:
values, tz = conversion.datetime_to_datetime64(arg)
dta = DatetimeArray(values, dtype=tz_to_dtype(tz))
return DatetimeIndex._simple_new(dta, name=name)
except (ValueError, TypeError):
raise e
if result is None:
assert format is None or infer_datetime_format
utc = tz == "utc"
result, tz_parsed = objects_to_datetime64ns(
arg,
dayfirst=dayfirst,
yearfirst=yearfirst,
utc=utc,
errors=errors,
require_iso8601=require_iso8601,
allow_object=True,
)
if tz_parsed is not None:
# We can take a shortcut since the datetime64 numpy array
# is in UTC
dta = DatetimeArray(result, dtype=tz_to_dtype(tz_parsed))
return DatetimeIndex._simple_new(dta, name=name)
utc = tz == "utc"
return _box_as_indexlike(result, utc=utc, name=name)
(
[pd.Period("2000", "D"), pd.Period("2001", "D")],
"Period[D]",
period_array(["2000", "2001"], freq="D"),
),
# Period dtype
(
[pd.Period("2000", "D")],
pd.PeriodDtype("D"),
period_array(["2000"], freq="D"),
),
# Datetime (naive)
(
[1, 2],
np.dtype("datetime64[ns]"),
DatetimeArray._from_sequence(np.array([1, 2], dtype="datetime64[ns]")),
),
(
np.array([1, 2], dtype="datetime64[ns]"),
None,
DatetimeArray._from_sequence(np.array([1, 2], dtype="datetime64[ns]")),
),
(
pd.DatetimeIndex(["2000", "2001"]),
np.dtype("datetime64[ns]"),
DatetimeArray._from_sequence(["2000", "2001"]),
),
(
pd.DatetimeIndex(["2000", "2001"]),
None,
DatetimeArray._from_sequence(["2000", "2001"]),
