Google Git
Sign in
apache / spark / 387a251a682a596ba4156b7d12e6025762ebac85 / . / python / pyspark / pandas / series.py
blob: 70f6a96cc78a75be0e48664a3604914c584d2ba1 [file] [log] [blame]
#
# Licensed to the Apache Software Foundation (ASF) under one or more
# contributor license agreements. See the NOTICE file distributed with
# this work for additional information regarding copyright ownership.
# The ASF licenses this file to You under the Apache License, Version 2.0
# (the "License"); you may not use this file except in compliance with
# the License. You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
"""
A wrapper class for Spark Column to behave similar to pandas Series.
"""
import datetime
import re
import inspect
import sys
from collections.abc import Mapping
from functools import partial, wraps, reduce
from typing import (
Any,
Callable,
Dict,
Generic,
IO,
Iterable,
List,
Optional,
Sequence,
Tuple,
Type,
Union,
cast,
no_type_check,
overload,
TYPE_CHECKING,
)
import numpy as np
import pandas as pd
from pandas.core.accessor import CachedAccessor
from pandas.io.formats.printing import pprint_thing
from pandas.api.types import is_list_like, is_hashable
from pandas.api.extensions import ExtensionDtype
from pandas.tseries.frequencies import DateOffset
from pyspark.sql import functions as F, Column, DataFrame as SparkDataFrame
from pyspark.sql.types import (
ArrayType,
BooleanType,
DataType,
DecimalType,
DoubleType,
FloatType,
IntegerType,
IntegralType,
LongType,
NumericType,
Row,
StructType,
)
from pyspark.sql.window import Window
from pyspark import pandas as ps # For running doctests and reference resolution in PyCharm.
from pyspark.pandas._typing import Axis, Dtype, Label, Name, Scalar, T
from pyspark.pandas.accessors import PandasOnSparkSeriesMethods
from pyspark.pandas.categorical import CategoricalAccessor
from pyspark.pandas.config import get_option
from pyspark.pandas.base import IndexOpsMixin
from pyspark.pandas.exceptions import SparkPandasIndexingError
from pyspark.pandas.frame import DataFrame
from pyspark.pandas.generic import Frame
from pyspark.pandas.internal import (
InternalField,
InternalFrame,
DEFAULT_SERIES_NAME,
NATURAL_ORDER_COLUMN_NAME,
SPARK_DEFAULT_INDEX_NAME,
SPARK_DEFAULT_SERIES_NAME,
)
from pyspark.pandas.missing.series import MissingPandasLikeSeries
from pyspark.pandas.plot import PandasOnSparkPlotAccessor
from pyspark.pandas.ml import corr
from pyspark.pandas.utils import (
combine_frames,
is_name_like_tuple,
is_name_like_value,
name_like_string,
same_anchor,
scol_for,
sql_conf,
validate_arguments_and_invoke_function,
validate_axis,
validate_bool_kwarg,
verify_temp_column_name,
SPARK_CONF_ARROW_ENABLED,
)
from pyspark.pandas.datetimes import DatetimeMethods
from pyspark.pandas.spark import functions as SF
from pyspark.pandas.spark.accessors import SparkSeriesMethods
from pyspark.pandas.strings import StringMethods
from pyspark.pandas.typedef import (
infer_return_type,
spark_type_to_pandas_dtype,
ScalarType,
SeriesType,
)
if TYPE_CHECKING:
from pyspark.sql._typing import ColumnOrName # noqa: F401 (SPARK-34943)
from pyspark.pandas.groupby import SeriesGroupBy # noqa: F401 (SPARK-34943)
from pyspark.pandas.indexes import Index # noqa: F401 (SPARK-34943)
# This regular expression pattern is complied and defined here to avoid to compile the same
# pattern every time it is used in _repr_ in Series.
# This pattern basically seeks the footer string from pandas'
REPR_PATTERN = re.compile(r"Length: (?P<length>[0-9]+)")
_flex_doc_SERIES = """
Return {desc} of series and other, element-wise (binary operator `{op_name}`).
Equivalent to ``{equiv}``
Parameters
----------
other : Series or scalar value
Returns
-------
Series
The result of the operation.
See Also
--------
Series.{reverse}
{series_examples}
"""
_add_example_SERIES = """
Examples
--------
>>> df = ps.DataFrame({'a': [2, 2, 4, np.nan],
... 'b': [2, np.nan, 2, np.nan]},
... index=['a', 'b', 'c', 'd'], columns=['a', 'b'])
>>> df
a b
a 2.0 2.0
b 2.0 NaN
c 4.0 2.0
d NaN NaN
>>> df.a.add(df.b)
a 4.0
b NaN
c 6.0
d NaN
dtype: float64
>>> df.a.radd(df.b)
a 4.0
b NaN
c 6.0
d NaN
dtype: float64
"""
_sub_example_SERIES = """
Examples
--------
>>> df = ps.DataFrame({'a': [2, 2, 4, np.nan],
... 'b': [2, np.nan, 2, np.nan]},
... index=['a', 'b', 'c', 'd'], columns=['a', 'b'])
>>> df
a b
a 2.0 2.0
b 2.0 NaN
c 4.0 2.0
d NaN NaN
>>> df.a.subtract(df.b)
a 0.0
b NaN
c 2.0
d NaN
dtype: float64
>>> df.a.rsub(df.b)
a 0.0
b NaN
c -2.0
d NaN
dtype: float64
"""
_mul_example_SERIES = """
Examples
--------
>>> df = ps.DataFrame({'a': [2, 2, 4, np.nan],
... 'b': [2, np.nan, 2, np.nan]},
... index=['a', 'b', 'c', 'd'], columns=['a', 'b'])
>>> df
a b
a 2.0 2.0
b 2.0 NaN
c 4.0 2.0
d NaN NaN
>>> df.a.multiply(df.b)
a 4.0
b NaN
c 8.0
d NaN
dtype: float64
>>> df.a.rmul(df.b)
a 4.0
b NaN
c 8.0
d NaN
dtype: float64
"""
_div_example_SERIES = """
Examples
--------
>>> df = ps.DataFrame({'a': [2, 2, 4, np.nan],
... 'b': [2, np.nan, 2, np.nan]},
... index=['a', 'b', 'c', 'd'], columns=['a', 'b'])
>>> df
a b
a 2.0 2.0
b 2.0 NaN
c 4.0 2.0
d NaN NaN
>>> df.a.divide(df.b)
a 1.0
b NaN
c 2.0
d NaN
dtype: float64
>>> df.a.rdiv(df.b)
a 1.0
b NaN
c 0.5
d NaN
dtype: float64
"""
_pow_example_SERIES = """
Examples
--------
>>> df = ps.DataFrame({'a': [2, 2, 4, np.nan],
... 'b': [2, np.nan, 2, np.nan]},
... index=['a', 'b', 'c', 'd'], columns=['a', 'b'])
>>> df
a b
a 2.0 2.0
b 2.0 NaN
c 4.0 2.0
d NaN NaN
>>> df.a.pow(df.b)
a 4.0
b NaN
c 16.0
d NaN
dtype: float64
>>> df.a.rpow(df.b)
a 4.0
b NaN
c 16.0
d NaN
dtype: float64
"""
_mod_example_SERIES = """
Examples
--------
>>> df = ps.DataFrame({'a': [2, 2, 4, np.nan],
... 'b': [2, np.nan, 2, np.nan]},
... index=['a', 'b', 'c', 'd'], columns=['a', 'b'])
>>> df
a b
a 2.0 2.0
b 2.0 NaN
c 4.0 2.0
d NaN NaN
>>> df.a.mod(df.b)
a 0.0
b NaN
c 0.0
d NaN
dtype: float64
>>> df.a.rmod(df.b)
a 0.0
b NaN
c 2.0
d NaN
dtype: float64
"""
_floordiv_example_SERIES = """
Examples
--------
>>> df = ps.DataFrame({'a': [2, 2, 4, np.nan],
... 'b': [2, np.nan, 2, np.nan]},
... index=['a', 'b', 'c', 'd'], columns=['a', 'b'])
>>> df
a b
a 2.0 2.0
b 2.0 NaN
c 4.0 2.0
d NaN NaN
>>> df.a.floordiv(df.b)
a 1.0
b NaN
c 2.0
d NaN
dtype: float64
>>> df.a.rfloordiv(df.b)
a 1.0
b NaN
c 0.0
d NaN
dtype: float64
"""
# Needed to disambiguate Series.str and str type
str_type = str
def _create_type_for_series_type(param: Any) -> Type[SeriesType]:
from pyspark.pandas.typedef import NameTypeHolder
if isinstance(param, ExtensionDtype):
new_class = type("NameType", (NameTypeHolder,), {}) # type: Type[NameTypeHolder]
new_class.tpe = param
else:
new_class = param.type if isinstance(param, np.dtype) else param
return SeriesType[new_class] # type: ignore
if (3, 5) <= sys.version_info < (3, 7) and __name__ != "__main__":
from typing import GenericMeta # type: ignore
old_getitem = GenericMeta.__getitem__ # type: ignore
@no_type_check
def new_getitem(self, params):
if hasattr(self, "is_series"):
return old_getitem(self, _create_type_for_series_type(params))
else:
return old_getitem(self, params)
GenericMeta.__getitem__ = new_getitem # type: ignore
class Series(Frame, IndexOpsMixin, Generic[T]):
"""
pandas-on-Spark Series that corresponds to pandas Series logically. This holds Spark Column
internally.
:ivar _internal: an internal immutable Frame to manage metadata.
:type _internal: InternalFrame
:ivar _psdf: Parent's pandas-on-Spark DataFrame
:type _psdf: ps.DataFrame
Parameters
----------
data : array-like, dict, or scalar value, pandas Series
Contains data stored in Series
If data is a dict, argument order is maintained for Python 3.6
and later.
Note that if `data` is a pandas Series, other arguments should not be used.
index : array-like or Index (1d)
Values must be hashable and have the same length as `data`.
Non-unique index values are allowed. Will default to
RangeIndex (0, 1, 2, ..., n) if not provided. If both a dict and index
sequence are used, the index will override the keys found in the
dict.
dtype : numpy.dtype or None
If None, dtype will be inferred
copy : boolean, default False
Copy input data
"""
@no_type_check
def __init__(self, data=None, index=None, dtype=None, name=None, copy=False, fastpath=False):
assert data is not None
if isinstance(data, DataFrame):
assert dtype is None
assert name is None
assert not copy
assert not fastpath
self._anchor = data # type: DataFrame
self._col_label = index # type: Label
else:
if isinstance(data, pd.Series):
assert index is None
assert dtype is None
assert name is None
assert not copy
assert not fastpath
s = data
else:
s = pd.Series(
data=data, index=index, dtype=dtype, name=name, copy=copy, fastpath=fastpath
)
internal = InternalFrame.from_pandas(pd.DataFrame(s))
if s.name is None:
internal = internal.copy(column_labels=[None])
anchor = DataFrame(internal)
self._anchor = anchor
self._col_label = anchor._internal.column_labels[0]
object.__setattr__(anchor, "_psseries", {self._column_label: self})
@property
def _psdf(self) -> DataFrame:
return self._anchor
@property
def _internal(self) -> InternalFrame:
return self._psdf._internal.select_column(self._column_label)
@property
def _column_label(self) -> Optional[Label]:
return self._col_label
def _update_anchor(self, psdf: DataFrame) -> None:
assert psdf._internal.column_labels == [self._column_label], (
psdf._internal.column_labels,
[self._column_label],
)
self._anchor = psdf
object.__setattr__(psdf, "_psseries", {self._column_label: self})
def _with_new_scol(self, scol: Column, *, field: Optional[InternalField] = None) -> "Series":
"""
Copy pandas-on-Spark Series with the new Spark Column.
:param scol: the new Spark Column
:return: the copied Series
"""
name = name_like_string(self._column_label)
internal = self._internal.copy(
data_spark_columns=[scol.alias(name)],
data_fields=[
field if field is None or field.struct_field is None else field.copy(name=name)
],
)
return first_series(DataFrame(internal))
spark = CachedAccessor("spark", SparkSeriesMethods)
@property
def dtypes(self) -> Dtype:
"""Return the dtype object of the underlying data.
>>> s = ps.Series(list('abc'))
>>> s.dtype == s.dtypes
True
"""
return self.dtype
@property
def axes(self) -> List["Index"]:
"""
Return a list of the row axis labels.
Examples
--------
>>> psser = ps.Series([1, 2, 3])
>>> psser.axes
[Int64Index([0, 1, 2], dtype='int64')]
"""
return [self.index]
# Arithmetic Operators
def add(self, other: Any) -> "Series":
return self + other
add.__doc__ = _flex_doc_SERIES.format(
desc="Addition",
op_name="+",
equiv="series + other",
reverse="radd",
series_examples=_add_example_SERIES,
)
def radd(self, other: Any) -> "Series":
return other + self
radd.__doc__ = _flex_doc_SERIES.format(
desc="Reverse Addition",
op_name="+",
equiv="other + series",
reverse="add",
series_examples=_add_example_SERIES,
)
def div(self, other: Any) -> "Series":
return self / other
div.__doc__ = _flex_doc_SERIES.format(
desc="Floating division",
op_name="/",
equiv="series / other",
reverse="rdiv",
series_examples=_div_example_SERIES,
)
divide = div
def rdiv(self, other: Any) -> "Series":
return other / self
rdiv.__doc__ = _flex_doc_SERIES.format(
desc="Reverse Floating division",
op_name="/",
equiv="other / series",
reverse="div",
series_examples=_div_example_SERIES,
)
def truediv(self, other: Any) -> "Series":
return self / other
truediv.__doc__ = _flex_doc_SERIES.format(
desc="Floating division",
op_name="/",
equiv="series / other",
reverse="rtruediv",
series_examples=_div_example_SERIES,
)
def rtruediv(self, other: Any) -> "Series":
return other / self
rtruediv.__doc__ = _flex_doc_SERIES.format(
desc="Reverse Floating division",
op_name="/",
equiv="other / series",
reverse="truediv",
series_examples=_div_example_SERIES,
)
def mul(self, other: Any) -> "Series":
return self * other
mul.__doc__ = _flex_doc_SERIES.format(
desc="Multiplication",
op_name="*",
equiv="series * other",
reverse="rmul",
series_examples=_mul_example_SERIES,
)
multiply = mul
def rmul(self, other: Any) -> "Series":
return other * self
rmul.__doc__ = _flex_doc_SERIES.format(
desc="Reverse Multiplication",
op_name="*",
equiv="other * series",
reverse="mul",
series_examples=_mul_example_SERIES,
)
def sub(self, other: Any) -> "Series":
return self - other
sub.__doc__ = _flex_doc_SERIES.format(
desc="Subtraction",
op_name="-",
equiv="series - other",
reverse="rsub",
series_examples=_sub_example_SERIES,
)
subtract = sub
def rsub(self, other: Any) -> "Series":
return other - self
rsub.__doc__ = _flex_doc_SERIES.format(
desc="Reverse Subtraction",
op_name="-",
equiv="other - series",
reverse="sub",
series_examples=_sub_example_SERIES,
)
def mod(self, other: Any) -> "Series":
return self % other
mod.__doc__ = _flex_doc_SERIES.format(
desc="Modulo",
op_name="%",
equiv="series % other",
reverse="rmod",
series_examples=_mod_example_SERIES,
)
def rmod(self, other: Any) -> "Series":
return other % self
rmod.__doc__ = _flex_doc_SERIES.format(
desc="Reverse Modulo",
op_name="%",
equiv="other % series",
reverse="mod",
series_examples=_mod_example_SERIES,
)
def pow(self, other: Any) -> "Series":
return self ** other
pow.__doc__ = _flex_doc_SERIES.format(
desc="Exponential power of series",
op_name="**",
equiv="series ** other",
reverse="rpow",
series_examples=_pow_example_SERIES,
)
def rpow(self, other: Any) -> "Series":
return other ** self
rpow.__doc__ = _flex_doc_SERIES.format(
desc="Reverse Exponential power",
op_name="**",
equiv="other ** series",
reverse="pow",
series_examples=_pow_example_SERIES,
)
def floordiv(self, other: Any) -> "Series":
return self // other
floordiv.__doc__ = _flex_doc_SERIES.format(
desc="Integer division",
op_name="//",
equiv="series // other",
reverse="rfloordiv",
series_examples=_floordiv_example_SERIES,
)
def rfloordiv(self, other: Any) -> "Series":
return other // self
rfloordiv.__doc__ = _flex_doc_SERIES.format(
desc="Reverse Integer division",
op_name="//",
equiv="other // series",
reverse="floordiv",
series_examples=_floordiv_example_SERIES,
)
# create accessor for pandas-on-Spark specific methods.
pandas_on_spark = CachedAccessor("pandas_on_spark", PandasOnSparkSeriesMethods)
# keep the name "koalas" for backward compatibility.
koalas = CachedAccessor("koalas", PandasOnSparkSeriesMethods)
# Comparison Operators
def eq(self, other: Any) -> bool:
"""
Compare if the current value is equal to the other.
>>> df = ps.DataFrame({'a': [1, 2, 3, 4],
... 'b': [1, np.nan, 1, np.nan]},
... index=['a', 'b', 'c', 'd'], columns=['a', 'b'])
>>> df.a == 1
a True
b False
c False
d False
Name: a, dtype: bool
>>> df.b.eq(1)
a True
b False
c True
d False
Name: b, dtype: bool
"""
return self == other
equals = eq
def gt(self, other: Any) -> "Series":
"""
Compare if the current value is greater than the other.
>>> df = ps.DataFrame({'a': [1, 2, 3, 4],
... 'b': [1, np.nan, 1, np.nan]},
... index=['a', 'b', 'c', 'd'], columns=['a', 'b'])
>>> df.a > 1
a False
b True
c True
d True
Name: a, dtype: bool
>>> df.b.gt(1)
a False
b False
c False
d False
Name: b, dtype: bool
"""
return self > other
def ge(self, other: Any) -> "Series":
"""
Compare if the current value is greater than or equal to the other.
>>> df = ps.DataFrame({'a': [1, 2, 3, 4],
... 'b': [1, np.nan, 1, np.nan]},
... index=['a', 'b', 'c', 'd'], columns=['a', 'b'])
>>> df.a >= 2
a False
b True
c True
d True
Name: a, dtype: bool
>>> df.b.ge(2)
a False
b False
c False
d False
Name: b, dtype: bool
"""
return self >= other
def lt(self, other: Any) -> "Series":
"""
Compare if the current value is less than the other.
>>> df = ps.DataFrame({'a': [1, 2, 3, 4],
... 'b': [1, np.nan, 1, np.nan]},
... index=['a', 'b', 'c', 'd'], columns=['a', 'b'])
>>> df.a < 1
a False
b False
c False
d False
Name: a, dtype: bool
>>> df.b.lt(2)
a True
b False
c True
d False
Name: b, dtype: bool
"""
return self < other
def le(self, other: Any) -> "Series":
"""
Compare if the current value is less than or equal to the other.
>>> df = ps.DataFrame({'a': [1, 2, 3, 4],
... 'b': [1, np.nan, 1, np.nan]},
... index=['a', 'b', 'c', 'd'], columns=['a', 'b'])
>>> df.a <= 2
a True
b True
c False
d False
Name: a, dtype: bool
>>> df.b.le(2)
a True
b False
c True
d False
Name: b, dtype: bool
"""
return self <= other
def ne(self, other: Any) -> "Series":
"""
Compare if the current value is not equal to the other.
>>> df = ps.DataFrame({'a': [1, 2, 3, 4],
... 'b': [1, np.nan, 1, np.nan]},
... index=['a', 'b', 'c', 'd'], columns=['a', 'b'])
>>> df.a != 1
a False
b True
c True
d True
Name: a, dtype: bool
>>> df.b.ne(1)
a False
b True
c False
d True
Name: b, dtype: bool
"""
return self != other
def divmod(self, other: Any) -> Tuple["Series", "Series"]:
"""
Return Integer division and modulo of series and other, element-wise
(binary operator `divmod`).
Parameters
----------
other : Series or scalar value
Returns
-------
2-Tuple of Series
The result of the operation.
See Also
--------
Series.rdivmod
"""
return self.floordiv(other), self.mod(other)
def rdivmod(self, other: Any) -> Tuple["Series", "Series"]:
"""
Return Integer division and modulo of series and other, element-wise
(binary operator `rdivmod`).
Parameters
----------
other : Series or scalar value
Returns
-------
2-Tuple of Series
The result of the operation.
See Also
--------
Series.divmod
"""
return self.rfloordiv(other), self.rmod(other)
def between(self, left: Any, right: Any, inclusive: bool = True) -> "Series":
"""
Return boolean Series equivalent to left <= series <= right.
This function returns a boolean vector containing `True` wherever the
corresponding Series element is between the boundary values `left` and
`right`. NA values are treated as `False`.
Parameters
----------
left : scalar or list-like
Left boundary.
right : scalar or list-like
Right boundary.
inclusive : bool, default True
Include boundaries.
Returns
-------
Series
Series representing whether each element is between left and
right (inclusive).
See Also
--------
Series.gt : Greater than of series and other.
Series.lt : Less than of series and other.
Notes
-----
This function is equivalent to ``(left <= ser) & (ser <= right)``
Examples
--------
>>> s = ps.Series([2, 0, 4, 8, np.nan])
Boundary values are included by default:
>>> s.between(1, 4)
0 True
1 False
2 True
3 False
4 False
dtype: bool
With `inclusive` set to ``False`` boundary values are excluded:
>>> s.between(1, 4, inclusive=False)
0 True
1 False
2 False
3 False
4 False
dtype: bool
`left` and `right` can be any scalar value:
>>> s = ps.Series(['Alice', 'Bob', 'Carol', 'Eve'])
>>> s.between('Anna', 'Daniel')
0 False
1 True
2 True
3 False
dtype: bool
"""
if inclusive:
lmask = self >= left
rmask = self <= right
else:
lmask = self > left
rmask = self < right
return lmask & rmask
# TODO: arg should support Series
# TODO: NaN and None
def map(self, arg: Union[Dict, Callable]) -> "Series":
"""
Map values of Series according to input correspondence.
Used for substituting each value in a Series with another value,
that may be derived from a function, a ``dict``.
.. note:: make sure the size of the dictionary is not huge because it could
downgrade the performance or throw OutOfMemoryError due to a huge
expression within Spark. Consider the input as a functions as an
alternative instead in this case.
Parameters
----------
arg : function or dict
Mapping correspondence.
Returns
-------
Series
Same index as caller.
See Also
--------
Series.apply : For applying more complex functions on a Series.
DataFrame.applymap : Apply a function elementwise on a whole DataFrame.
Notes
-----
When ``arg`` is a dictionary, values in Series that are not in the
dictionary (as keys) are converted to ``None``. However, if the
dictionary is a ``dict`` subclass that defines ``__missing__`` (i.e.
provides a method for default values), then this default is used
rather than ``None``.
Examples
--------
>>> s = ps.Series(['cat', 'dog', None, 'rabbit'])
>>> s
0 cat
1 dog
2 None
3 rabbit
dtype: object
``map`` accepts a ``dict``. Values that are not found
in the ``dict`` are converted to ``None``, unless the dict has a default
value (e.g. ``defaultdict``):
>>> s.map({'cat': 'kitten', 'dog': 'puppy'})
0 kitten
1 puppy
2 None
3 None
dtype: object
It also accepts a function:
>>> def format(x) -> str:
... return 'I am a {}'.format(x)
>>> s.map(format)
0 I am a cat
1 I am a dog
2 I am a None
3 I am a rabbit
dtype: object
"""
if isinstance(arg, dict):
is_start = True
# In case dictionary is empty.
current = F.when(SF.lit(False), SF.lit(None).cast(self.spark.data_type))
for to_replace, value in arg.items():
if is_start:
current = F.when(self.spark.column == SF.lit(to_replace), value)
is_start = False
else:
current = current.when(self.spark.column == SF.lit(to_replace), value)
if hasattr(arg, "__missing__"):
tmp_val = arg[np._NoValue] # type: ignore
# Remove in case it's set in defaultdict.
del arg[np._NoValue] # type: ignore
current = current.otherwise(SF.lit(tmp_val))
else:
current = current.otherwise(SF.lit(None).cast(self.spark.data_type))
return self._with_new_scol(current)
else:
return self.apply(arg)
@property
def shape(self) -> Tuple[int]:
"""Return a tuple of the shape of the underlying data."""
return (len(self),)
@property
def name(self) -> Name:
"""Return name of the Series."""
name = self._column_label
if name is not None and len(name) == 1:
return name[0]
else:
return name
@name.setter
def name(self, name: Name) -> None:
self.rename(name, inplace=True)
# TODO: Functionality and documentation should be matched. Currently, changing index labels
# taking dictionary and function to change index are not supported.
def rename(self, index: Optional[Name] = None, **kwargs: Any) -> "Series":
"""
Alter Series name.
Parameters
----------
index : scalar
Scalar will alter the ``Series.name`` attribute.
inplace : bool, default False
Whether to return a new Series. If True then value of copy is
ignored.
Returns
-------
Series
Series with name altered.
Examples
--------
>>> s = ps.Series([1, 2, 3])
>>> s
0 1
1 2
2 3
dtype: int64
>>> s.rename("my_name") # scalar, changes Series.name
0 1
1 2
2 3
Name: my_name, dtype: int64
"""
if index is None:
pass
elif not is_hashable(index):
raise TypeError("Series.name must be a hashable type")
elif not isinstance(index, tuple):
index = (index,)
name = name_like_string(index)
scol = self.spark.column.alias(name)
field = self._internal.data_fields[0].copy(name=name)
internal = self._internal.copy(
column_labels=[index],
data_spark_columns=[scol],
data_fields=[field],
column_label_names=None,
)
psdf = DataFrame(internal) # type: DataFrame
if kwargs.get("inplace", False):
self._col_label = index
self._update_anchor(psdf)
return self
else:
return first_series(psdf)
def rename_axis(
self, mapper: Optional[Any] = None, index: Optional[Any] = None, inplace: bool = False
) -> Optional["Series"]:
"""
Set the name of the axis for the index or columns.
Parameters
----------
mapper, index : scalar, list-like, dict-like or function, optional
A scalar, list-like, dict-like or functions transformations to
apply to the index values.
inplace : bool, default False
Modifies the object directly, instead of creating a new Series.
Returns
-------
Series, or None if `inplace` is True.
See Also
--------
Series.rename : Alter Series index labels or name.
DataFrame.rename : Alter DataFrame index labels or name.
Index.rename : Set new names on index.
Examples
--------
>>> s = ps.Series(["dog", "cat", "monkey"], name="animal")
>>> s # doctest: +NORMALIZE_WHITESPACE
0 dog
1 cat
2 monkey
Name: animal, dtype: object
>>> s.rename_axis("index").sort_index() # doctest: +NORMALIZE_WHITESPACE
index
0 dog
1 cat
2 monkey
Name: animal, dtype: object
**MultiIndex**
>>> index = pd.MultiIndex.from_product([['mammal'],
... ['dog', 'cat', 'monkey']],
... names=['type', 'name'])
>>> s = ps.Series([4, 4, 2], index=index, name='num_legs')
>>> s # doctest: +NORMALIZE_WHITESPACE
type name
mammal dog 4
cat 4
monkey 2
Name: num_legs, dtype: int64
>>> s.rename_axis(index={'type': 'class'}).sort_index() # doctest: +NORMALIZE_WHITESPACE
class name
mammal cat 4
dog 4
monkey 2
Name: num_legs, dtype: int64
>>> s.rename_axis(index=str.upper).sort_index() # doctest: +NORMALIZE_WHITESPACE
TYPE NAME
mammal cat 4
dog 4
monkey 2
Name: num_legs, dtype: int64
"""
psdf = self.to_frame().rename_axis(mapper=mapper, index=index, inplace=False)
if inplace:
self._update_anchor(psdf)
return None
else:
return first_series(psdf)
@property
def index(self) -> "ps.Index":
"""The index (axis labels) Column of the Series.
See Also
--------
Index
"""
return self._psdf.index
@property
def is_unique(self) -> bool:
"""
Return boolean if values in the object are unique
Returns
-------
is_unique : boolean
>>> ps.Series([1, 2, 3]).is_unique
True
>>> ps.Series([1, 2, 2]).is_unique
False
>>> ps.Series([1, 2, 3, None]).is_unique
True
"""
scol = self.spark.column
# Here we check:
# 1. the distinct count without nulls and count without nulls for non-null values
# 2. count null values and see if null is a distinct value.
#
# This workaround is in order to calculate the distinct count including nulls in
# single pass. Note that COUNT(DISTINCT expr) in Spark is designed to ignore nulls.
return self._internal.spark_frame.select(
(F.count(scol) == F.countDistinct(scol))
& (F.count(F.when(scol.isNull(), 1).otherwise(None)) <= 1)
).collect()[0][0]
def reset_index(
self,
level: Optional[Union[int, Name, Sequence[Union[int, Name]]]] = None,
drop: bool = False,
name: Optional[Name] = None,
inplace: bool = False,
) -> Optional[Union["Series", DataFrame]]:
"""
Generate a new DataFrame or Series with the index reset.
This is useful when the index needs to be treated as a column,
or when the index is meaningless and needs to be reset
to the default before another operation.
Parameters
----------
level : int, str, tuple, or list, default optional
For a Series with a MultiIndex, only remove the specified levels from the index.
Removes all levels by default.
drop : bool, default False
Just reset the index, without inserting it as a column in the new DataFrame.
name : object, optional
The name to use for the column containing the original Series values.
Uses self.name by default. This argument is ignored when drop is True.
inplace : bool, default False
Modify the Series in place (do not create a new object).
Returns
-------
Series or DataFrame
When `drop` is False (the default), a DataFrame is returned.
The newly created columns will come first in the DataFrame,
followed by the original Series values.
When `drop` is True, a `Series` is returned.
In either case, if ``inplace=True``, no value is returned.
Examples
--------
>>> s = ps.Series([1, 2, 3, 4], index=pd.Index(['a', 'b', 'c', 'd'], name='idx'))
Generate a DataFrame with default index.
>>> s.reset_index()
idx 0
0 a 1
1 b 2
2 c 3
3 d 4
To specify the name of the new column use `name`.
>>> s.reset_index(name='values')
idx values
0 a 1
1 b 2
2 c 3
3 d 4
To generate a new Series with the default set `drop` to True.
>>> s.reset_index(drop=True)
0 1
1 2
2 3
3 4
dtype: int64
To update the Series in place, without generating a new one
set `inplace` to True. Note that it also requires ``drop=True``.
>>> s.reset_index(inplace=True, drop=True)
>>> s
0 1
1 2
2 3
3 4
dtype: int64
"""
inplace = validate_bool_kwarg(inplace, "inplace")
if inplace and not drop:
raise TypeError("Cannot reset_index inplace on a Series to create a DataFrame")
if drop:
psdf = self._psdf[[self.name]]
else:
psser = self
if name is not None:
psser = psser.rename(name)
psdf = psser.to_frame()
psdf = psdf.reset_index(level=level, drop=drop)
if drop:
if inplace:
self._update_anchor(psdf)
return None
else:
return first_series(psdf)
else:
return psdf
def to_frame(self, name: Optional[Name] = None) -> DataFrame:
"""
Convert Series to DataFrame.
Parameters
----------
name : object, default None
The passed name should substitute for the series name (if it has
one).
Returns
-------
DataFrame
DataFrame representation of Series.
Examples
--------
>>> s = ps.Series(["a", "b", "c"])
>>> s.to_frame()
0
0 a
1 b
2 c
>>> s = ps.Series(["a", "b", "c"], name="vals")
>>> s.to_frame()
vals
0 a
1 b
2 c
"""
if name is not None:
renamed = self.rename(name)
elif self._column_label is None:
renamed = self.rename(DEFAULT_SERIES_NAME)
else:
renamed = self
return DataFrame(renamed._internal)
to_dataframe = to_frame
def to_string(
self,
buf: Optional[IO[str]] = None,
na_rep: str = "NaN",
float_format: Optional[Callable[[float], str]] = None,
header: bool = True,
index: bool = True,
length: bool = False,
dtype: bool = False,
name: bool = False,
max_rows: Optional[int] = None,
) -> Optional[str]:
"""
Render a string representation of the Series.
.. note:: This method should only be used if the resulting pandas object is expected
to be small, as all the data is loaded into the driver's memory. If the input
is large, set max_rows parameter.
Parameters
----------
buf : StringIO-like, optional
buffer to write to
na_rep : string, optional
string representation of NAN to use, default 'NaN'
float_format : one-parameter function, optional
formatter function to apply to columns' elements if they are floats
default None
header : boolean, default True
Add the Series header (index name)
index : bool, optional
Add index (row) labels, default True
length : boolean, default False
Add the Series length
dtype : boolean, default False
Add the Series dtype
name : boolean, default False
Add the Series name if not None
max_rows : int, optional
Maximum number of rows to show before truncating. If None, show
all.
Returns
-------
formatted : string (if not buffer passed)
Examples
--------
>>> df = ps.DataFrame([(.2, .3), (.0, .6), (.6, .0), (.2, .1)], columns=['dogs', 'cats'])
>>> print(df['dogs'].to_string())
0 0.2
1 0.0
2 0.6
3 0.2
>>> print(df['dogs'].to_string(max_rows=2))
0 0.2
1 0.0
"""
# Make sure locals() call is at the top of the function so we don't capture local variables.
args = locals()
if max_rows is not None:
psseries = self.head(max_rows)
else:
psseries = self
return validate_arguments_and_invoke_function(
psseries._to_internal_pandas(), self.to_string, pd.Series.to_string, args
)
def to_clipboard(self, excel: bool = True, sep: Optional[str] = None, **kwargs: Any) -> None:
# Docstring defined below by reusing DataFrame.to_clipboard's.
args = locals()
psseries = self
return validate_arguments_and_invoke_function(
psseries._to_internal_pandas(), self.to_clipboard, pd.Series.to_clipboard, args
)
to_clipboard.__doc__ = DataFrame.to_clipboard.__doc__
def to_dict(self, into: Type = dict) -> Mapping:
"""
Convert Series to {label -> value} dict or dict-like object.
.. note:: This method should only be used if the resulting pandas DataFrame is expected
to be small, as all the data is loaded into the driver's memory.
Parameters
----------
into : class, default dict
The collections.abc.Mapping subclass to use as the return
object. Can be the actual class or an empty
instance of the mapping type you want. If you want a
collections.defaultdict, you must pass it initialized.
Returns
-------
collections.abc.Mapping
Key-value representation of Series.
Examples
--------
>>> s = ps.Series([1, 2, 3, 4])
>>> s_dict = s.to_dict()
>>> sorted(s_dict.items())
[(0, 1), (1, 2), (2, 3), (3, 4)]
>>> from collections import OrderedDict, defaultdict
>>> s.to_dict(OrderedDict)
OrderedDict([(0, 1), (1, 2), (2, 3), (3, 4)])
>>> dd = defaultdict(list)
>>> s.to_dict(dd) # doctest: +ELLIPSIS
defaultdict(<class 'list'>, {...})
"""
# Make sure locals() call is at the top of the function so we don't capture local variables.
args = locals()
psseries = self
return validate_arguments_and_invoke_function(
psseries._to_internal_pandas(), self.to_dict, pd.Series.to_dict, args
)
def to_latex(
self,
buf: Optional[IO[str]] = None,
columns: Optional[List[Name]] = None,
col_space: Optional[int] = None,
header: bool = True,
index: bool = True,
na_rep: str = "NaN",
formatters: Optional[
Union[List[Callable[[Any], str]], Dict[Name, Callable[[Any], str]]]
] = None,
float_format: Optional[Callable[[float], str]] = None,
sparsify: Optional[bool] = None,
index_names: bool = True,
bold_rows: bool = False,
column_format: Optional[str] = None,
longtable: Optional[bool] = None,
escape: Optional[bool] = None,
encoding: Optional[str] = None,
decimal: str = ".",
multicolumn: Optional[bool] = None,
multicolumn_format: Optional[str] = None,
multirow: Optional[bool] = None,
) -> Optional[str]:
args = locals()
psseries = self
return validate_arguments_and_invoke_function(
psseries._to_internal_pandas(), self.to_latex, pd.Series.to_latex, args
)
to_latex.__doc__ = DataFrame.to_latex.__doc__
def to_pandas(self) -> pd.Series:
"""
Return a pandas Series.
.. note:: This method should only be used if the resulting pandas object is expected
to be small, as all the data is loaded into the driver's memory.
Examples
--------
>>> df = ps.DataFrame([(.2, .3), (.0, .6), (.6, .0), (.2, .1)], columns=['dogs', 'cats'])
>>> df['dogs'].to_pandas()
0 0.2
1 0.0
2 0.6
3 0.2
Name: dogs, dtype: float64
"""
return self._to_internal_pandas().copy()
def to_list(self) -> List:
"""
Return a list of the values.
These are each a scalar type, which is a Python scalar
(for str, int, float) or a pandas scalar
(for Timestamp/Timedelta/Interval/Period)
.. note:: This method should only be used if the resulting list is expected
to be small, as all the data is loaded into the driver's memory.
"""
return self._to_internal_pandas().tolist()
tolist = to_list
def drop_duplicates(self, keep: str = "first", inplace: bool = False) -> Optional["Series"]:
"""
Return Series with duplicate values removed.
Parameters
----------
keep : {'first', 'last', ``False``}, default 'first'
Method to handle dropping duplicates:
- 'first' : Drop duplicates except for the first occurrence.
- 'last' : Drop duplicates except for the last occurrence.
- ``False`` : Drop all duplicates.
inplace : bool, default ``False``
If ``True``, performs operation inplace and returns None.
Returns
-------
Series
Series with duplicates dropped.
Examples
--------
Generate a Series with duplicated entries.
>>> s = ps.Series(['lama', 'cow', 'lama', 'beetle', 'lama', 'hippo'],
... name='animal')
>>> s.sort_index()
0 lama
1 cow
2 lama
3 beetle
4 lama
5 hippo
Name: animal, dtype: object
With the 'keep' parameter, the selection behaviour of duplicated values
can be changed. The value 'first' keeps the first occurrence for each
set of duplicated entries. The default value of keep is 'first'.
>>> s.drop_duplicates().sort_index()
0 lama
1 cow
3 beetle
5 hippo
Name: animal, dtype: object
The value 'last' for parameter 'keep' keeps the last occurrence for
each set of duplicated entries.
>>> s.drop_duplicates(keep='last').sort_index()
1 cow
3 beetle
4 lama
5 hippo
Name: animal, dtype: object
The value ``False`` for parameter 'keep' discards all sets of
duplicated entries. Setting the value of 'inplace' to ``True`` performs
the operation inplace and returns ``None``.
>>> s.drop_duplicates(keep=False, inplace=True)
>>> s.sort_index()
1 cow
3 beetle
5 hippo
Name: animal, dtype: object
"""
inplace = validate_bool_kwarg(inplace, "inplace")
psdf = self._psdf[[self.name]].drop_duplicates(keep=keep)
if inplace:
self._update_anchor(psdf)
return None
else:
return first_series(psdf)
def reindex(self, index: Optional[Any] = None, fill_value: Optional[Any] = None) -> "Series":
"""
Conform Series to new index with optional filling logic, placing
NA/NaN in locations having no value in the previous index. A new object
is produced.
Parameters
----------
index: array-like, optional
New labels / index to conform to, should be specified using keywords.
Preferably an Index object to avoid duplicating data
fill_value : scalar, default np.NaN
Value to use for missing values. Defaults to NaN, but can be any
"compatible" value.
Returns
-------
Series with changed index.
See Also
--------
Series.reset_index : Remove row labels or move them to new columns.
Examples
--------
Create a series with some fictional data.
>>> index = ['Firefox', 'Chrome', 'Safari', 'IE10', 'Konqueror']
>>> ser = ps.Series([200, 200, 404, 404, 301],
... index=index, name='http_status')
>>> ser
Firefox 200
Chrome 200
Safari 404
IE10 404
Konqueror 301
Name: http_status, dtype: int64
Create a new index and reindex the Series. By default
values in the new index that do not have corresponding
records in the Series are assigned ``NaN``.
>>> new_index= ['Safari', 'Iceweasel', 'Comodo Dragon', 'IE10',
... 'Chrome']
>>> ser.reindex(new_index).sort_index()
Chrome 200.0
Comodo Dragon NaN
IE10 404.0
Iceweasel NaN
Safari 404.0
Name: http_status, dtype: float64
We can fill in the missing values by passing a value to
the keyword ``fill_value``.
>>> ser.reindex(new_index, fill_value=0).sort_index()
Chrome 200
Comodo Dragon 0
IE10 404
Iceweasel 0
Safari 404
Name: http_status, dtype: int64
To further illustrate the filling functionality in
``reindex``, we will create a Series with a
monotonically increasing index (for example, a sequence
of dates).
>>> date_index = pd.date_range('1/1/2010', periods=6, freq='D')
>>> ser2 = ps.Series([100, 101, np.nan, 100, 89, 88],
... name='prices', index=date_index)
>>> ser2.sort_index()
2010-01-01 100.0
2010-01-02 101.0
2010-01-03 NaN
2010-01-04 100.0
2010-01-05 89.0
2010-01-06 88.0
Name: prices, dtype: float64
Suppose we decide to expand the series to cover a wider
date range.
>>> date_index2 = pd.date_range('12/29/2009', periods=10, freq='D')
>>> ser2.reindex(date_index2).sort_index()
2009-12-29 NaN
2009-12-30 NaN
2009-12-31 NaN
2010-01-01 100.0
2010-01-02 101.0
2010-01-03 NaN
2010-01-04 100.0
2010-01-05 89.0
2010-01-06 88.0
2010-01-07 NaN
Name: prices, dtype: float64
"""
return first_series(self.to_frame().reindex(index=index, fill_value=fill_value)).rename(
self.name
)
def reindex_like(self, other: Union["Series", "DataFrame"]) -> "Series":
"""
Return a Series with matching indices as other object.
Conform the object to the same index on all axes. Places NA/NaN in locations
having no value in the previous index.
Parameters
----------
other : Series or DataFrame
Its row and column indices are used to define the new indices
of this object.
Returns
-------
Series
Series with changed indices on each axis.
See Also
--------
DataFrame.set_index : Set row labels.
DataFrame.reset_index : Remove row labels or move them to new columns.
DataFrame.reindex : Change to new indices or expand indices.
Notes
-----
Same as calling
``.reindex(index=other.index, ...)``.
Examples
--------
>>> s1 = ps.Series([24.3, 31.0, 22.0, 35.0],
... index=pd.date_range(start='2014-02-12',
... end='2014-02-15', freq='D'),
... name="temp_celsius")
>>> s1
2014-02-12 24.3
2014-02-13 31.0
2014-02-14 22.0
2014-02-15 35.0
Name: temp_celsius, dtype: float64
>>> s2 = ps.Series(["low", "low", "medium"],
... index=pd.DatetimeIndex(['2014-02-12', '2014-02-13',
... '2014-02-15']),
... name="winspeed")
>>> s2
2014-02-12 low
2014-02-13 low
2014-02-15 medium
Name: winspeed, dtype: object
>>> s2.reindex_like(s1).sort_index()
2014-02-12 low
2014-02-13 low
2014-02-14 None
2014-02-15 medium
Name: winspeed, dtype: object
"""
if isinstance(other, (Series, DataFrame)):
return self.reindex(index=other.index)
else:
raise TypeError("other must be a pandas-on-Spark Series or DataFrame")
def fillna(
self,
value: Optional[Any] = None,
method: Optional[str] = None,
axis: Optional[Axis] = None,
inplace: bool = False,
limit: Optional[int] = None,
) -> Optional["Series"]:
"""Fill NA/NaN values.
.. note:: the current implementation of 'method' parameter in fillna uses Spark's Window
without specifying partition specification. This leads to move all data into
single partition in single machine and could cause serious
performance degradation. Avoid this method against very large dataset.
Parameters
----------
value : scalar, dict, Series
Value to use to fill holes. alternately a dict/Series of values
specifying which value to use for each column.
DataFrame is not supported.
method : {'backfill', 'bfill', 'pad', 'ffill', None}, default None
Method to use for filling holes in reindexed Series pad / ffill: propagate last valid
observation forward to next valid backfill / bfill:
use NEXT valid observation to fill gap
axis : {0 or `index`}
1 and `columns` are not supported.
inplace : boolean, default False
Fill in place (do not create a new object)
limit : int, default None
If method is specified, this is the maximum number of consecutive NaN values to
forward/backward fill. In other words, if there is a gap with more than this number of
consecutive NaNs, it will only be partially filled. If method is not specified,
this is the maximum number of entries along the entire axis where NaNs will be filled.
Must be greater than 0 if not None
Returns
-------
Series
Series with NA entries filled.
Examples
--------
>>> s = ps.Series([np.nan, 2, 3, 4, np.nan, 6], name='x')
>>> s
0 NaN
1 2.0
2 3.0
3 4.0
4 NaN
5 6.0
Name: x, dtype: float64
Replace all NaN elements with 0s.
>>> s.fillna(0)
0 0.0
1 2.0
2 3.0
3 4.0
4 0.0
5 6.0
Name: x, dtype: float64
We can also propagate non-null values forward or backward.
>>> s.fillna(method='ffill')
0 NaN
1 2.0
2 3.0
3 4.0
4 4.0
5 6.0
Name: x, dtype: float64
>>> s = ps.Series([np.nan, 'a', 'b', 'c', np.nan], name='x')
>>> s.fillna(method='ffill')
0 None
1 a
2 b
3 c
4 c
Name: x, dtype: object
"""
psser = self._fillna(value=value, method=method, axis=axis, limit=limit)
if method is not None:
psser = DataFrame(psser._psdf._internal.resolved_copy)._psser_for(self._column_label)
inplace = validate_bool_kwarg(inplace, "inplace")
if inplace:
self._psdf._update_internal_frame(psser._psdf._internal, requires_same_anchor=False)
return None
else:
return psser.copy()
def _fillna(
self,
value: Optional[Any] = None,
method: Optional[str] = None,
axis: Optional[Axis] = None,
limit: Optional[int] = None,
part_cols: Sequence["ColumnOrName"] = (),
) -> "Series":
axis = validate_axis(axis)
if axis != 0:
raise NotImplementedError("fillna currently only works for axis=0 or axis='index'")
if (value is None) and (method is None):
raise ValueError("Must specify a fillna 'value' or 'method' parameter.")
if (method is not None) and (method not in ["ffill", "pad", "backfill", "bfill"]):
raise ValueError("Expecting 'pad', 'ffill', 'backfill' or 'bfill'.")
scol = self.spark.column
if not self.spark.nullable and not isinstance(
self.spark.data_type, (FloatType, DoubleType)
):
return self._psdf.copy()._psser_for(self._column_label)
cond = self.isnull().spark.column
if value is not None:
if not isinstance(value, (float, int, str, bool)):
raise TypeError("Unsupported type %s" % type(value).__name__)
if limit is not None:
raise ValueError("limit parameter for value is not support now")
scol = F.when(cond, value).otherwise(scol)
else:
if method in ["ffill", "pad"]:
func = F.last
end = Window.currentRow - 1
if limit is not None:
begin = Window.currentRow - limit
else:
begin = Window.unboundedPreceding
elif method in ["bfill", "backfill"]:
func = F.first
begin = Window.currentRow + 1
if limit is not None:
end = Window.currentRow + limit
else:
end = Window.unboundedFollowing
window = (
Window.partitionBy(*part_cols)
.orderBy(NATURAL_ORDER_COLUMN_NAME)
.rowsBetween(begin, end)
)
scol = F.when(cond, func(scol, True).over(window)).otherwise(scol)
return DataFrame(
self._psdf._internal.with_new_spark_column(
self._column_label, scol.alias(name_like_string(self.name)) # TODO: dtype?
)
)._psser_for(self._column_label)
def dropna(self, axis: Axis = 0, inplace: bool = False, **kwargs: Any) -> Optional["Series"]:
"""
Return a new Series with missing values removed.
Parameters
----------
axis : {0 or 'index'}, default 0
There is only one axis to drop values from.
inplace : bool, default False
If True, do operation inplace and return None.
**kwargs
Not in use.
Returns
-------
Series
Series with NA entries dropped from it.
Examples
--------
>>> ser = ps.Series([1., 2., np.nan])
>>> ser
0 1.0
1 2.0
2 NaN
dtype: float64
Drop NA values from a Series.
>>> ser.dropna()
0 1.0
1 2.0
dtype: float64
Keep the Series with valid entries in the same variable.
>>> ser.dropna(inplace=True)
>>> ser
0 1.0
1 2.0
dtype: float64
"""
inplace = validate_bool_kwarg(inplace, "inplace")
# TODO: last two examples from pandas produce different results.
psdf = self._psdf[[self.name]].dropna(axis=axis, inplace=False)
if inplace:
self._update_anchor(psdf)
return None
else:
return first_series(psdf)
def clip(self, lower: Union[float, int] = None, upper: Union[float, int] = None) -> "Series":
"""
Trim values at input threshold(s).
Assigns values outside boundary to boundary values.
Parameters
----------
lower : float or int, default None
Minimum threshold value. All values below this threshold will be set to it.
upper : float or int, default None
Maximum threshold value. All values above this threshold will be set to it.
Returns
-------
Series
Series with the values outside the clip boundaries replaced
Examples
--------
>>> ps.Series([0, 2, 4]).clip(1, 3)
0 1
1 2
2 3
dtype: int64
Notes
-----
One difference between this implementation and pandas is that running
`pd.Series(['a', 'b']).clip(0, 1)` will crash with "TypeError: '<=' not supported between
instances of 'str' and 'int'" while `ps.Series(['a', 'b']).clip(0, 1)` will output the
original Series, simply ignoring the incompatible types.
"""
if is_list_like(lower) or is_list_like(upper):
raise TypeError(
"List-like value are not supported for 'lower' and 'upper' at the " + "moment"
)
if lower is None and upper is None:
return self
if isinstance(self.spark.data_type, NumericType):
scol = self.spark.column
if lower is not None:
scol = F.when(scol < lower, lower).otherwise(scol)
if upper is not None:
scol = F.when(scol > upper, upper).otherwise(scol)
return self._with_new_scol(
scol.alias(self._internal.data_spark_column_names[0]),
field=self._internal.data_fields[0],
)
else:
return self
def drop(
self,
labels: Optional[Union[Name, List[Name]]] = None,
index: Optional[Union[Name, List[Name]]] = None,
level: Optional[int] = None,
) -> "Series":
"""
Return Series with specified index labels removed.
Remove elements of a Series based on specifying the index labels.
When using a multi-index, labels on different levels can be removed by specifying the level.
Parameters
----------
labels : single label or list-like
Index labels to drop.
index : None
Redundant for application on Series, but index can be used instead of labels.
level : int or level name, optional
For MultiIndex, level for which the labels will be removed.
Returns
-------
Series
Series with specified index labels removed.
See Also
--------
Series.dropna
Examples
--------
>>> s = ps.Series(data=np.arange(3), index=['A', 'B', 'C'])
>>> s
A 0
B 1
C 2
dtype: int64
Drop single label A
>>> s.drop('A')
B 1
C 2
dtype: int64
Drop labels B and C
>>> s.drop(labels=['B', 'C'])
A 0
dtype: int64
With 'index' rather than 'labels' returns exactly same result.
>>> s.drop(index='A')
B 1
C 2
dtype: int64
>>> s.drop(index=['B', 'C'])
A 0
dtype: int64
Also support for MultiIndex
>>> midx = pd.MultiIndex([['lama', 'cow', 'falcon'],
... ['speed', 'weight', 'length']],
... [[0, 0, 0, 1, 1, 1, 2, 2, 2],
... [0, 1, 2, 0, 1, 2, 0, 1, 2]])
>>> s = ps.Series([45, 200, 1.2, 30, 250, 1.5, 320, 1, 0.3],
... index=midx)
>>> s
lama speed 45.0
weight 200.0
length 1.2
cow speed 30.0
weight 250.0
length 1.5
falcon speed 320.0
weight 1.0
length 0.3
dtype: float64
>>> s.drop(labels='weight', level=1)
lama speed 45.0
length 1.2
cow speed 30.0
length 1.5
falcon speed 320.0
length 0.3
dtype: float64
>>> s.drop(('lama', 'weight'))
lama speed 45.0
length 1.2
cow speed 30.0
weight 250.0
length 1.5
falcon speed 320.0
weight 1.0
length 0.3
dtype: float64
>>> s.drop([('lama', 'speed'), ('falcon', 'weight')])
lama weight 200.0
length 1.2
cow speed 30.0
weight 250.0
length 1.5
falcon speed 320.0
length 0.3
dtype: float64
"""
return first_series(self._drop(labels=labels, index=index, level=level))
def _drop(
self,
labels: Optional[Union[Name, List[Name]]] = None,
index: Optional[Union[Name, List[Name]]] = None,
level: Optional[int] = None,
) -> DataFrame:
if labels is not None:
if index is not None:
raise ValueError("Cannot specify both 'labels' and 'index'")
return self._drop(index=labels, level=level)
if index is not None:
internal = self._internal
if level is None:
level = 0
if level >= internal.index_level:
raise ValueError("'level' should be less than the number of indexes")
if is_name_like_tuple(index): # type: ignore
index_list = [cast(Label, index)]
elif is_name_like_value(index):
index_list = [(index,)]
elif all(is_name_like_value(idxes, allow_tuple=False) for idxes in index):
index_list = [(idex,) for idex in index]
elif not all(is_name_like_tuple(idxes) for idxes in index):
raise ValueError(
"If the given index is a list, it "
"should only contains names as all tuples or all non tuples "
"that contain index names"
)
else:
index_list = cast(List[Label], index)
drop_index_scols = []
for idxes in index_list:
try:
index_scols = [
internal.index_spark_columns[lvl] == idx
for lvl, idx in enumerate(idxes, level)
]
except IndexError:
raise KeyError(
"Key length ({}) exceeds index depth ({})".format(
internal.index_level, len(idxes)
)
)
drop_index_scols.append(reduce(lambda x, y: x & y, index_scols))
cond = ~reduce(lambda x, y: x | y, drop_index_scols)
return DataFrame(internal.with_filter(cond))
else:
raise ValueError("Need to specify at least one of 'labels' or 'index'")
def head(self, n: int = 5) -> "Series":
"""
Return the first n rows.
This function returns the first n rows for the object based on position.
It is useful for quickly testing if your object has the right type of data in it.
Parameters
----------
n : Integer, default = 5
Returns
-------
The first n rows of the caller object.
Examples
--------
>>> df = ps.DataFrame({'animal':['alligator', 'bee', 'falcon', 'lion']})
>>> df.animal.head(2) # doctest: +NORMALIZE_WHITESPACE
0 alligator
1 bee
Name: animal, dtype: object
"""
return first_series(self.to_frame().head(n)).rename(self.name)
def last(self, offset: Union[str, DateOffset]) -> "Series":
"""
Select final periods of time series data based on a date offset.
When having a Series with dates as index, this function can
select the last few elements based on a date offset.
Parameters
----------
offset : str or DateOffset
The offset length of the data that will be selected. For instance,
'3D' will display all the rows having their index within the last 3 days.
Returns
-------
Series
A subset of the caller.
Raises
------
TypeError
If the index is not a :class:`DatetimeIndex`
Examples
--------
>>> index = pd.date_range('2018-04-09', periods=4, freq='2D')
>>> psser = ps.Series([1, 2, 3, 4], index=index)
>>> psser
2018-04-09 1
2018-04-11 2
2018-04-13 3
2018-04-15 4
dtype: int64
Get the rows for the last 3 days:
>>> psser.last('3D')
2018-04-13 3
2018-04-15 4
dtype: int64
Notice the data for 3 last calendar days were returned, not the last
3 observed days in the dataset, and therefore data for 2018-04-11 was
not returned.
"""
return first_series(self.to_frame().last(offset)).rename(self.name)
def first(self, offset: Union[str, DateOffset]) -> "Series":
"""
Select first periods of time series data based on a date offset.
When having a Series with dates as index, this function can
select the first few elements based on a date offset.
Parameters
----------
offset : str or DateOffset
The offset length of the data that will be selected. For instance,
'3D' will display all the rows having their index within the first 3 days.
Returns
-------
Series
A subset of the caller.
Raises
------
TypeError
If the index is not a :class:`DatetimeIndex`
Examples
--------
>>> index = pd.date_range('2018-04-09', periods=4, freq='2D')
>>> psser = ps.Series([1, 2, 3, 4], index=index)
>>> psser
2018-04-09 1
2018-04-11 2
2018-04-13 3
2018-04-15 4
dtype: int64
Get the rows for the first 3 days:
>>> psser.first('3D')
2018-04-09 1
2018-04-11 2
dtype: int64
Notice the data for 3 first calendar days were returned, not the first
3 observed days in the dataset, and therefore data for 2018-04-13 was
not returned.
"""
return first_series(self.to_frame().first(offset)).rename(self.name)
# TODO: Categorical type isn't supported (due to PySpark's limitation) and
# some doctests related with timestamps were not added.
def unique(self) -> "Series":
"""
Return unique values of Series object.
Uniques are returned in order of appearance. Hash table-based unique,
therefore does NOT sort.
.. note:: This method returns newly created Series whereas pandas returns
the unique values as a NumPy array.
Returns
-------
Returns the unique values as a Series.
See Also
--------
Index.unique
groupby.SeriesGroupBy.unique
Examples
--------
>>> psser = ps.Series([2, 1, 3, 3], name='A')
>>> psser.unique().sort_values() # doctest: +NORMALIZE_WHITESPACE, +ELLIPSIS
<BLANKLINE>
... 1
... 2
... 3
Name: A, dtype: int64
>>> ps.Series([pd.Timestamp('2016-01-01') for _ in range(3)]).unique()
0 2016-01-01
dtype: datetime64[ns]
>>> psser.name = ('x', 'a')
>>> psser.unique().sort_values() # doctest: +NORMALIZE_WHITESPACE, +ELLIPSIS
<BLANKLINE>
... 1
... 2
... 3
Name: (x, a), dtype: int64
"""
sdf = self._internal.spark_frame.select(self.spark.column).distinct()
internal = InternalFrame(
spark_frame=sdf,
index_spark_columns=None,
column_labels=[self._column_label],
data_spark_columns=[scol_for(sdf, self._internal.data_spark_column_names[0])],
data_fields=[self._internal.data_fields[0]],
column_label_names=self._internal.column_label_names,
)
return first_series(DataFrame(internal))
def sort_values(
self, ascending: bool = True, inplace: bool = False, na_position: str = "last"
) -> Optional["Series"]:
"""
Sort by the values.
Sort a Series in ascending or descending order by some criterion.
Parameters
----------
ascending : bool or list of bool, default True
Sort ascending vs. descending. Specify list for multiple sort
orders. If this is a list of bools, must match the length of
the by.
inplace : bool, default False
if True, perform operation in-place
na_position : {'first', 'last'}, default 'last'
`first` puts NaNs at the beginning, `last` puts NaNs at the end
Returns
-------
sorted_obj : Series ordered by values.
Examples
--------
>>> s = ps.Series([np.nan, 1, 3, 10, 5])
>>> s
0 NaN
1 1.0
2 3.0
3 10.0
4 5.0
dtype: float64
Sort values ascending order (default behaviour)
>>> s.sort_values(ascending=True)
1 1.0
2 3.0
4 5.0
3 10.0
0 NaN
dtype: float64
Sort values descending order
>>> s.sort_values(ascending=False)
3 10.0
4 5.0
2 3.0
1 1.0
0 NaN
dtype: float64
Sort values inplace
>>> s.sort_values(ascending=False, inplace=True)
>>> s
3 10.0
4 5.0
2 3.0
1 1.0
0 NaN
dtype: float64
Sort values putting NAs first
>>> s.sort_values(na_position='first')
0 NaN
1 1.0
2 3.0
4 5.0
3 10.0
dtype: float64
Sort a series of strings
>>> s = ps.Series(['z', 'b', 'd', 'a', 'c'])
>>> s
0 z
1 b
2 d
3 a
4 c
dtype: object
>>> s.sort_values()
3 a
1 b
4 c
2 d
0 z
dtype: object
"""
inplace = validate_bool_kwarg(inplace, "inplace")
psdf = self._psdf[[self.name]]._sort(
by=[self.spark.column], ascending=ascending, na_position=na_position
)
if inplace:
self._update_anchor(psdf)
return None
else:
return first_series(psdf)
def sort_index(
self,
axis: Axis = 0,
level: Optional[Union[int, List[int]]] = None,
ascending: bool = True,
inplace: bool = False,
kind: str = None,
na_position: str = "last",
) -> Optional["Series"]:
"""
Sort object by labels (along an axis)
Parameters
----------
axis : index, columns to direct sorting. Currently, only axis = 0 is supported.
level : int or level name or list of ints or list of level names
if not None, sort on values in specified index level(s)
ascending : boolean, default True
Sort ascending vs. descending
inplace : bool, default False
if True, perform operation in-place
kind : str, default None
pandas-on-Spark does not allow specifying the sorting algorithm at the moment,
default None
na_position : {‘first’, ‘last’}, default ‘last’
first puts NaNs at the beginning, last puts NaNs at the end. Not implemented for
MultiIndex.
Returns
-------
sorted_obj : Series
Examples
--------
>>> df = ps.Series([2, 1, np.nan], index=['b', 'a', np.nan])
>>> df.sort_index()
a 1.0
b 2.0
NaN NaN
dtype: float64
>>> df.sort_index(ascending=False)
b 2.0
a 1.0
NaN NaN
dtype: float64
>>> df.sort_index(na_position='first')
NaN NaN
a 1.0
b 2.0
dtype: float64
>>> df.sort_index(inplace=True)
>>> df
a 1.0
b 2.0
NaN NaN
dtype: float64
>>> df = ps.Series(range(4), index=[['b', 'b', 'a', 'a'], [1, 0, 1, 0]], name='0')
>>> df.sort_index()
a 0 3
1 2
b 0 1
1 0
Name: 0, dtype: int64
>>> df.sort_index(level=1) # doctest: +SKIP
a 0 3
b 0 1
a 1 2
b 1 0
Name: 0, dtype: int64
>>> df.sort_index(level=[1, 0])
a 0 3
b 0 1
a 1 2
b 1 0
Name: 0, dtype: int64
"""
inplace = validate_bool_kwarg(inplace, "inplace")
psdf = self._psdf[[self.name]].sort_index(
axis=axis, level=level, ascending=ascending, kind=kind, na_position=na_position
)
if inplace:
self._update_anchor(psdf)
return None
else:
return first_series(psdf)
def swaplevel(
self, i: Union[int, Name] = -2, j: Union[int, Name] = -1, copy: bool = True
) -> "Series":
"""
Swap levels i and j in a MultiIndex.
Default is to swap the two innermost levels of the index.
Parameters
----------
i, j : int, str
Level of the indices to be swapped. Can pass level name as string.
copy : bool, default True
Whether to copy underlying data. Must be True.
Returns
-------
Series
Series with levels swapped in MultiIndex.
Examples
--------
>>> midx = pd.MultiIndex.from_arrays([['a', 'b'], [1, 2]], names = ['word', 'number'])
>>> midx # doctest: +SKIP
MultiIndex([('a', 1),
('b', 2)],
names=['word', 'number'])
>>> psser = ps.Series(['x', 'y'], index=midx)
>>> psser
word number
a 1 x
b 2 y
dtype: object
>>> psser.swaplevel()
number word
1 a x
2 b y
dtype: object
>>> psser.swaplevel(0, 1)
number word
1 a x
2 b y
dtype: object
>>> psser.swaplevel('number', 'word')
number word
1 a x
2 b y
dtype: object
"""
assert copy is True
return first_series(self.to_frame().swaplevel(i, j, axis=0)).rename(self.name)
def swapaxes(self, i: Axis, j: Axis, copy: bool = True) -> "Series":
"""
Interchange axes and swap values axes appropriately.
Parameters
----------
i: {0 or 'index', 1 or 'columns'}. The axis to swap.
j: {0 or 'index', 1 or 'columns'}. The axis to swap.
copy : bool, default True.
Returns
-------
Series
Examples
--------
>>> psser = ps.Series([1, 2, 3], index=["x", "y", "z"])
>>> psser
x 1
y 2
z 3
dtype: int64
>>>
>>> psser.swapaxes(0, 0)
x 1
y 2
z 3
dtype: int64
"""
assert copy is True
i = validate_axis(i)
j = validate_axis(j)
if not i == j == 0:
raise ValueError("Axis must be 0 for Series")
return self.copy()
def add_prefix(self, prefix: str) -> "Series":
"""
Prefix labels with string `prefix`.
For Series, the row labels are prefixed.
For DataFrame, the column labels are prefixed.
Parameters
----------
prefix : str
The string to add before each label.
Returns
-------
Series
New Series with updated labels.
See Also
--------
Series.add_suffix: Suffix column labels with string `suffix`.
DataFrame.add_suffix: Suffix column labels with string `suffix`.
DataFrame.add_prefix: Prefix column labels with string `prefix`.
Examples
--------
>>> s = ps.Series([1, 2, 3, 4])
>>> s
0 1
1 2
2 3
3 4
dtype: int64
>>> s.add_prefix('item_')
item_0 1
item_1 2
item_2 3
item_3 4
dtype: int64
"""
assert isinstance(prefix, str)
internal = self._internal.resolved_copy
sdf = internal.spark_frame.select(
[
F.concat(SF.lit(prefix), index_spark_column).alias(index_spark_column_name)
for index_spark_column, index_spark_column_name in zip(
internal.index_spark_columns, internal.index_spark_column_names
)
]
+ internal.data_spark_columns
)
return first_series(
DataFrame(internal.with_new_sdf(sdf, index_fields=([None] * internal.index_level)))
)
def add_suffix(self, suffix: str) -> "Series":
"""
Suffix labels with string suffix.
For Series, the row labels are suffixed.
For DataFrame, the column labels are suffixed.
Parameters
----------
suffix : str
The string to add after each label.
Returns
-------
Series
New Series with updated labels.
See Also
--------
Series.add_prefix: Prefix row labels with string `prefix`.
DataFrame.add_prefix: Prefix column labels with string `prefix`.
DataFrame.add_suffix: Suffix column labels with string `suffix`.
Examples
--------
>>> s = ps.Series([1, 2, 3, 4])
>>> s
0 1
1 2
2 3
3 4
dtype: int64
>>> s.add_suffix('_item')
0_item 1
1_item 2
2_item 3
3_item 4
dtype: int64
"""
assert isinstance(suffix, str)
internal = self._internal.resolved_copy
sdf = internal.spark_frame.select(
[
F.concat(index_spark_column, SF.lit(suffix)).alias(index_spark_column_name)
for index_spark_column, index_spark_column_name in zip(
internal.index_spark_columns, internal.index_spark_column_names
)
]
+ internal.data_spark_columns
)
return first_series(
DataFrame(internal.with_new_sdf(sdf, index_fields=([None] * internal.index_level)))
)
def corr(self, other: "Series", method: str = "pearson") -> float:
"""
Compute correlation with `other` Series, excluding missing values.
Parameters
----------
other : Series
method : {'pearson', 'spearman'}
* pearson : standard correlation coefficient
* spearman : Spearman rank correlation
Returns
-------
correlation : float
Examples
--------
>>> df = ps.DataFrame({'s1': [.2, .0, .6, .2],
... 's2': [.3, .6, .0, .1]})
>>> s1 = df.s1
>>> s2 = df.s2
>>> s1.corr(s2, method='pearson') # doctest: +ELLIPSIS
-0.851064...
>>> s1.corr(s2, method='spearman') # doctest: +ELLIPSIS
-0.948683...
Notes
-----
There are behavior differences between pandas-on-Spark and pandas.
* the `method` argument only accepts 'pearson', 'spearman'
* the data should not contain NaNs. pandas-on-Spark will return an error.
* pandas-on-Spark doesn't support the following argument(s).
* `min_periods` argument is not supported
"""
# This implementation is suboptimal because it computes more than necessary,
# but it should be a start
columns = ["__corr_arg1__", "__corr_arg2__"]
psdf = self._psdf.assign(__corr_arg1__=self, __corr_arg2__=other)[columns]
psdf.columns = columns
c = corr(psdf, method=method)
return c.loc[tuple(columns)]
def nsmallest(self, n: int = 5) -> "Series":
"""
Return the smallest `n` elements.
Parameters
----------
n : int, default 5
Return this many ascending sorted values.
Returns
-------
Series
The `n` smallest values in the Series, sorted in increasing order.
See Also
--------
Series.nlargest: Get the `n` largest elements.
Series.sort_values: Sort Series by values.
Series.head: Return the first `n` rows.
Notes
-----
Faster than ``.sort_values().head(n)`` for small `n` relative to
the size of the ``Series`` object.
In pandas-on-Spark, thanks to Spark's lazy execution and query optimizer,
the two would have same performance.
Examples
--------
>>> data = [1, 2, 3, 4, np.nan ,6, 7, 8]
>>> s = ps.Series(data)
>>> s
0 1.0
1 2.0
2 3.0
3 4.0
4 NaN
5 6.0
6 7.0
7 8.0
dtype: float64
The `n` largest elements where ``n=5`` by default.
>>> s.nsmallest()
0 1.0
1 2.0
2 3.0
3 4.0
5 6.0
dtype: float64
>>> s.nsmallest(3)
0 1.0
1 2.0
2 3.0
dtype: float64
"""
return self.sort_values(ascending=True).head(n)
def nlargest(self, n: int = 5) -> "Series":
"""
Return the largest `n` elements.
Parameters
----------
n : int, default 5
Returns
-------
Series
The `n` largest values in the Series, sorted in decreasing order.
See Also
--------
Series.nsmallest: Get the `n` smallest elements.
Series.sort_values: Sort Series by values.
Series.head: Return the first `n` rows.
Notes
-----
Faster than ``.sort_values(ascending=False).head(n)`` for small `n`
relative to the size of the ``Series`` object.
In pandas-on-Spark, thanks to Spark's lazy execution and query optimizer,
the two would have same performance.
Examples
--------
>>> data = [1, 2, 3, 4, np.nan ,6, 7, 8]
>>> s = ps.Series(data)
>>> s
0 1.0
1 2.0
2 3.0
3 4.0
4 NaN
5 6.0
6 7.0
7 8.0
dtype: float64
The `n` largest elements where ``n=5`` by default.
>>> s.nlargest()
7 8.0
6 7.0
5 6.0
3 4.0
2 3.0
dtype: float64
>>> s.nlargest(n=3)
7 8.0
6 7.0
5 6.0
dtype: float64
"""
return self.sort_values(ascending=False).head(n)
def append(
self, to_append: "Series", ignore_index: bool = False, verify_integrity: bool = False
) -> "Series":
"""
Concatenate two or more Series.
Parameters
----------
to_append : Series or list/tuple of Series
ignore_index : boolean, default False
If True, do not use the index labels.
verify_integrity : boolean, default False
If True, raise Exception on creating index with duplicates
Returns
-------
appended : Series
Examples
--------
>>> s1 = ps.Series([1, 2, 3])
>>> s2 = ps.Series([4, 5, 6])
>>> s3 = ps.Series([4, 5, 6], index=[3,4,5])
>>> s1.append(s2)
0 1
1 2
2 3
0 4
1 5
2 6
dtype: int64
>>> s1.append(s3)
0 1
1 2
2 3
3 4
4 5
5 6
dtype: int64
With ignore_index set to True:
>>> s1.append(s2, ignore_index=True)
0 1
1 2
2 3
3 4
4 5
5 6
dtype: int64
"""
return first_series(
self.to_frame().append(to_append.to_frame(), ignore_index, verify_integrity)
).rename(self.name)
def sample(
self,
n: Optional[int] = None,
frac: Optional[float] = None,
replace: bool = False,
random_state: Optional[int] = None,
) -> "Series":
return first_series(
self.to_frame().sample(n=n, frac=frac, replace=replace, random_state=random_state)
).rename(self.name)
sample.__doc__ = DataFrame.sample.__doc__
@no_type_check
def hist(self, bins=10, **kwds):
return self.plot.hist(bins, **kwds)
hist.__doc__ = PandasOnSparkPlotAccessor.hist.__doc__
def apply(self, func: Callable, args: Sequence[Any] = (), **kwds: Any) -> "Series":
"""
Invoke function on values of Series.
Can be a Python function that only works on the Series.
.. note:: this API executes the function once to infer the type which is
potentially expensive, for instance, when the dataset is created after
aggregations or sorting.
To avoid this, specify return type in ``func``, for instance, as below:
>>> def square(x) -> np.int32:
... return x ** 2
pandas-on-Spark uses return type hint and does not try to infer the type.
Parameters
----------
func : function
Python function to apply. Note that type hint for return type is required.
args : tuple
Positional arguments passed to func after the series value.
**kwds
Additional keyword arguments passed to func.
Returns
-------
Series
See Also
--------
Series.aggregate : Only perform aggregating type operations.
Series.transform : Only perform transforming type operations.
DataFrame.apply : The equivalent function for DataFrame.
Examples
--------
Create a Series with typical summer temperatures for each city.
>>> s = ps.Series([20, 21, 12],
... index=['London', 'New York', 'Helsinki'])
>>> s
London 20
New York 21
Helsinki 12
dtype: int64
Square the values by defining a function and passing it as an
argument to ``apply()``.
>>> def square(x) -> np.int64:
... return x ** 2
>>> s.apply(square)
London 400
New York 441
Helsinki 144
dtype: int64
Define a custom function that needs additional positional
arguments and pass these additional arguments using the
``args`` keyword
>>> def subtract_custom_value(x, custom_value) -> np.int64:
... return x - custom_value
>>> s.apply(subtract_custom_value, args=(5,))
London 15
New York 16
Helsinki 7
dtype: int64
Define a custom function that takes keyword arguments
and pass these arguments to ``apply``
>>> def add_custom_values(x, **kwargs) -> np.int64:
... for month in kwargs:
... x += kwargs[month]
... return x
>>> s.apply(add_custom_values, june=30, july=20, august=25)
London 95
New York 96
Helsinki 87
dtype: int64
Use a function from the Numpy library
>>> def numpy_log(col) -> np.float64:
... return np.log(col)
>>> s.apply(numpy_log)
London 2.995732
New York 3.044522
Helsinki 2.484907
dtype: float64
You can omit the type hint and let pandas-on-Spark infer its type.
>>> s.apply(np.log)
London 2.995732
New York 3.044522
Helsinki 2.484907
dtype: float64
"""
assert callable(func), "the first argument should be a callable function."
try:
spec = inspect.getfullargspec(func)
return_sig = spec.annotations.get("return", None)
should_infer_schema = return_sig is None
except TypeError:
# Falls back to schema inference if it fails to get signature.
should_infer_schema = True
apply_each = wraps(func)(lambda s: s.apply(func, args=args, **kwds))
if should_infer_schema:
return self.pandas_on_spark._transform_batch(apply_each, None)
else:
sig_return = infer_return_type(func)
if not isinstance(sig_return, ScalarType):
raise ValueError(
"Expected the return type of this function to be of scalar type, "
"but found type {}".format(sig_return)
)
return_type = cast(ScalarType, sig_return)
return self.pandas_on_spark._transform_batch(apply_each, return_type)
# TODO: not all arguments are implemented comparing to pandas' for now.
def aggregate(self, func: Union[str, List[str]]) -> Union[Scalar, "Series"]:
"""Aggregate using one or more operations over the specified axis.
Parameters
----------
func : str or a list of str
function name(s) as string apply to series.
Returns
-------
scalar, Series
The return can be:
- scalar : when Series.agg is called with single function
- Series : when Series.agg is called with several functions
Notes
-----
`agg` is an alias for `aggregate`. Use the alias.
See Also
--------
Series.apply : Invoke function on a Series.
Series.transform : Only perform transforming type operations.
Series.groupby : Perform operations over groups.
DataFrame.aggregate : The equivalent function for DataFrame.
Examples
--------
>>> s = ps.Series([1, 2, 3, 4])
>>> s.agg('min')
1
>>> s.agg(['min', 'max']).sort_index()
max 4
min 1
dtype: int64
"""
if isinstance(func, list):
return first_series(self.to_frame().aggregate(func)).rename(self.name)
elif isinstance(func, str):
return getattr(self, func)()
else:
raise TypeError("func must be a string or list of strings")
agg = aggregate
def transpose(self, *args: Any, **kwargs: Any) -> "Series":
"""
Return the transpose, which is by definition self.
Examples
--------
It returns the same object as the transpose of the given series object, which is by
definition self.
>>> s = ps.Series([1, 2, 3])
>>> s
0 1
1 2
2 3
dtype: int64
>>> s.transpose()
0 1
1 2
2 3
dtype: int64
"""
return self.copy()
T = property(transpose)
def transform(
self, func: Union[Callable, List[Callable]], axis: Axis = 0, *args: Any, **kwargs: Any
) -> Union["Series", DataFrame]:
"""
Call ``func`` producing the same type as `self` with transformed values
and that has the same axis length as input.
.. note:: this API executes the function once to infer the type which is
potentially expensive, for instance, when the dataset is created after
aggregations or sorting.
To avoid this, specify return type in ``func``, for instance, as below:
>>> def square(x) -> np.int32:
... return x ** 2
pandas-on-Spark uses return type hint and does not try to infer the type.
Parameters
----------
func : function or list
A function or a list of functions to use for transforming the data.
axis : int, default 0 or 'index'
Can only be set to 0 at the moment.
*args
Positional arguments to pass to `func`.
**kwargs
Keyword arguments to pass to `func`.
Returns
-------
An instance of the same type with `self` that must have the same length as input.
See Also
--------
Series.aggregate : Only perform aggregating type operations.
Series.apply : Invoke function on Series.
DataFrame.transform : The equivalent function for DataFrame.
Examples
--------
>>> s = ps.Series(range(3))
>>> s
0 0
1 1
2 2
dtype: int64
>>> def sqrt(x) -> float:
... return np.sqrt(x)
>>> s.transform(sqrt)
0 0.000000
1 1.000000
2 1.414214
dtype: float64
Even though the resulting instance must have the same length as the
input, it is possible to provide several input functions:
>>> def exp(x) -> float:
... return np.exp(x)
>>> s.transform([sqrt, exp])
sqrt exp
0 0.000000 1.000000
1 1.000000 2.718282
2 1.414214 7.389056
You can omit the type hint and let pandas-on-Spark infer its type.
>>> s.transform([np.sqrt, np.exp])
sqrt exp
0 0.000000 1.000000
1 1.000000 2.718282
2 1.414214 7.389056
"""
axis = validate_axis(axis)
if axis != 0:
raise NotImplementedError('axis should be either 0 or "index" currently.')
if isinstance(func, list):
applied = []
for f in func:
applied.append(self.apply(f, args=args, **kwargs).rename(f.__name__))
internal = self._internal.with_new_columns(applied)
return DataFrame(internal)
else:
return self.apply(func, args=args, **kwargs)
def round(self, decimals: int = 0) -> "Series":
"""
Round each value in a Series to the given number of decimals.
Parameters
----------
decimals : int
Number of decimal places to round to (default: 0).
If decimals is negative, it specifies the number of
positions to the left of the decimal point.
Returns
-------
Series object
See Also
--------
DataFrame.round
Examples
--------
>>> df = ps.Series([0.028208, 0.038683, 0.877076], name='x')
>>> df
0 0.028208
1 0.038683
2 0.877076
Name: x, dtype: float64
>>> df.round(2)
0 0.03
1 0.04
2 0.88
Name: x, dtype: float64
"""
if not isinstance(decimals, int):
raise TypeError("decimals must be an integer")
scol = F.round(self.spark.column, decimals)
return self._with_new_scol(
scol,
field=(
self._internal.data_fields[0].copy(nullable=True)
if not isinstance(self.spark.data_type, DecimalType)
else None
),
)
# TODO: add 'interpolation' parameter.
def quantile(
self, q: Union[float, Iterable[float]] = 0.5, accuracy: int = 10000
) -> Union[Scalar, "Series"]:
"""
Return value at the given quantile.
.. note:: Unlike pandas', the quantile in pandas-on-Spark is an approximated quantile
based upon approximate percentile computation because computing quantile across
a large dataset is extremely expensive.
Parameters
----------
q : float or array-like, default 0.5 (50% quantile)
0 <= q <= 1, the quantile(s) to compute.
accuracy : int, optional
Default accuracy of approximation. Larger value means better accuracy.
The relative error can be deduced by 1.0 / accuracy.
Returns
-------
float or Series
If the current object is a Series and ``q`` is an array, a Series will be
returned where the index is ``q`` and the values are the quantiles, otherwise
a float will be returned.
Examples
--------
>>> s = ps.Series([1, 2, 3, 4, 5])
>>> s.quantile(.5)
3.0
>>> (s + 1).quantile(.5)
4.0
>>> s.quantile([.25, .5, .75])
0.25 2.0
0.50 3.0
0.75 4.0
dtype: float64
>>> (s + 1).quantile([.25, .5, .75])
0.25 3.0
0.50 4.0
0.75 5.0
dtype: float64
"""
if isinstance(q, Iterable):
return first_series(
self.to_frame().quantile(q=q, axis=0, numeric_only=False, accuracy=accuracy)
).rename(self.name)
else:
if not isinstance(accuracy, int):
raise TypeError(
"accuracy must be an integer; however, got [%s]" % type(accuracy).__name__
)
if not isinstance(q, float):
raise TypeError(
"q must be a float or an array of floats; however, [%s] found." % type(q)
)
q_float = cast(float, q)
if q_float < 0.0 or q_float > 1.0:
raise ValueError("percentiles should all be in the interval [0, 1].")
def quantile(spark_column: Column, spark_type: DataType) -> Column:
if isinstance(spark_type, (BooleanType, NumericType)):
return F.percentile_approx(spark_column.cast(DoubleType()), q_float, accuracy)
else:
raise TypeError(
"Could not convert {} ({}) to numeric".format(
spark_type_to_pandas_dtype(spark_type), spark_type.simpleString()
)
)
return self._reduce_for_stat_function(quantile, name="quantile")
# TODO: add axis, numeric_only, pct, na_option parameter
def rank(self, method: str = "average", ascending: bool = True) -> "Series":
"""
Compute numerical data ranks (1 through n) along axis. Equal values are
assigned a rank that is the average of the ranks of those values.
.. note:: the current implementation of rank uses Spark's Window without
specifying partition specification. This leads to move all data into
single partition in single machine and could cause serious
performance degradation. Avoid this method against very large dataset.
Parameters
----------
method : {'average', 'min', 'max', 'first', 'dense'}
* average: average rank of group
* min: lowest rank in group
* max: highest rank in group
* first: ranks assigned in order they appear in the array
* dense: like 'min', but rank always increases by 1 between groups
ascending : boolean, default True
False for ranks by high (1) to low (N)
Returns
-------
ranks : same type as caller
Examples
--------
>>> s = ps.Series([1, 2, 2, 3], name='A')
>>> s
0 1
1 2
2 2
3 3
Name: A, dtype: int64
>>> s.rank()
0 1.0
1 2.5
2 2.5
3 4.0
Name: A, dtype: float64
If method is set to 'min', it use lowest rank in group.
>>> s.rank(method='min')
0 1.0
1 2.0
2 2.0
3 4.0
Name: A, dtype: float64
If method is set to 'max', it use highest rank in group.
>>> s.rank(method='max')
0 1.0
1 3.0
2 3.0
3 4.0
Name: A, dtype: float64
If method is set to 'first', it is assigned rank in order without groups.
>>> s.rank(method='first')
0 1.0
1 2.0
2 3.0
3 4.0
Name: A, dtype: float64
If method is set to 'dense', it leaves no gaps in group.
>>> s.rank(method='dense')
0 1.0
1 2.0
2 2.0
3 3.0
Name: A, dtype: float64
"""
return self._rank(method, ascending).spark.analyzed
def _rank(
self,
method: str = "average",
ascending: bool = True,
*,
part_cols: Sequence["ColumnOrName"] = ()
) -> "Series":
if method not in ["average", "min", "max", "first", "dense"]:
msg = "method must be one of 'average', 'min', 'max', 'first', 'dense'"
raise ValueError(msg)
if self._internal.index_level > 1:
raise ValueError("rank do not support index now")
if ascending:
asc_func = lambda scol: scol.asc()
else:
asc_func = lambda scol: scol.desc()
if method == "first":
window = (
Window.orderBy(
asc_func(self.spark.column),
asc_func(F.col(NATURAL_ORDER_COLUMN_NAME)),
)
.partitionBy(*part_cols)
.rowsBetween(Window.unboundedPreceding, Window.currentRow)
)
scol = F.row_number().over(window)
elif method == "dense":
window = (
Window.orderBy(asc_func(self.spark.column))
.partitionBy(*part_cols)
.rowsBetween(Window.unboundedPreceding, Window.currentRow)
)
scol = F.dense_rank().over(window)
else:
if method == "average":
stat_func = F.mean
elif method == "min":
stat_func = F.min
elif method == "max":
stat_func = F.max
window1 = (
Window.orderBy(asc_func(self.spark.column))
.partitionBy(*part_cols)
.rowsBetween(Window.unboundedPreceding, Window.currentRow)
)
window2 = Window.partitionBy([self.spark.column] + list(part_cols)).rowsBetween(
Window.unboundedPreceding, Window.unboundedFollowing
)
scol = stat_func(F.row_number().over(window1)).over(window2)
return self._with_new_scol(scol.cast(DoubleType()))
def filter(
self,
items: Optional[Sequence[Any]] = None,
like: Optional[str] = None,
regex: Optional[str] = None,
axis: Optional[Axis] = None,
) -> "Series":
axis = validate_axis(axis)
if axis == 1:
raise ValueError("Series does not support columns axis.")
return first_series(
self.to_frame().filter(items=items, like=like, regex=regex, axis=axis)
).rename(self.name)
filter.__doc__ = DataFrame.filter.__doc__
def describe(self, percentiles: Optional[List[float]] = None) -> "Series":
return first_series(self.to_frame().describe(percentiles)).rename(self.name)
describe.__doc__ = DataFrame.describe.__doc__
def diff(self, periods: int = 1) -> "Series":
"""
First discrete difference of element.
Calculates the difference of a Series element compared with another element in the
DataFrame (default is the element in the same column of the previous row).
.. note:: the current implementation of diff uses Spark's Window without
specifying partition specification. This leads to move all data into
single partition in single machine and could cause serious
performance degradation. Avoid this method against very large dataset.
Parameters
----------
periods : int, default 1
Periods to shift for calculating difference, accepts negative values.
Returns
-------
diffed : Series
Examples
--------
>>> df = ps.DataFrame({'a': [1, 2, 3, 4, 5, 6],
... 'b': [1, 1, 2, 3, 5, 8],
... 'c': [1, 4, 9, 16, 25, 36]}, columns=['a', 'b', 'c'])
>>> df
a b c
0 1 1 1
1 2 1 4
2 3 2 9
3 4 3 16
4 5 5 25
5 6 8 36
>>> df.b.diff()
0 NaN
1 0.0
2 1.0
3 1.0
4 2.0
5 3.0
Name: b, dtype: float64
Difference with previous value
>>> df.c.diff(periods=3)
0 NaN
1 NaN
2 NaN
3 15.0
4 21.0
5 27.0
Name: c, dtype: float64
Difference with following value
>>> df.c.diff(periods=-1)
0 -3.0
1 -5.0
2 -7.0
3 -9.0
4 -11.0
5 NaN
Name: c, dtype: float64
"""
return self._diff(periods).spark.analyzed
def _diff(self, periods: int, *, part_cols: Sequence["ColumnOrName"] = ()) -> "Series":
if not isinstance(periods, int):
raise TypeError("periods should be an int; however, got [%s]" % type(periods).__name__)
window = (
Window.partitionBy(*part_cols)
.orderBy(NATURAL_ORDER_COLUMN_NAME)
.rowsBetween(-periods, -periods)
)
scol = self.spark.column - F.lag(self.spark.column, periods).over(window)
return self._with_new_scol(scol, field=self._internal.data_fields[0].copy(nullable=True))
def idxmax(self, skipna: bool = True) -> Union[Tuple, Any]:
"""
Return the row label of the maximum value.
If multiple values equal the maximum, the first row label with that
value is returned.
Parameters
----------
skipna : bool, default True
Exclude NA/null values. If the entire Series is NA, the result
will be NA.
Returns
-------
Index
Label of the maximum value.
Raises
------
ValueError
If the Series is empty.
See Also
--------
Series.idxmin : Return index *label* of the first occurrence
of minimum of values.
Examples
--------
>>> s = ps.Series(data=[1, None, 4, 3, 5],
... index=['A', 'B', 'C', 'D', 'E'])
>>> s
A 1.0
B NaN
C 4.0
D 3.0
E 5.0
dtype: float64
>>> s.idxmax()
'E'
If `skipna` is False and there is an NA value in the data,
the function returns ``nan``.
>>> s.idxmax(skipna=False)
nan
In case of multi-index, you get a tuple:
>>> index = pd.MultiIndex.from_arrays([
... ['a', 'a', 'b', 'b'], ['c', 'd', 'e', 'f']], names=('first', 'second'))
>>> s = ps.Series(data=[1, None, 4, 5], index=index)
>>> s
first second
a c 1.0
d NaN
b e 4.0
f 5.0
dtype: float64
>>> s.idxmax()
('b', 'f')
If multiple values equal the maximum, the first row label with that
value is returned.
>>> s = ps.Series([1, 100, 1, 100, 1, 100], index=[10, 3, 5, 2, 1, 8])
>>> s
10 1
3 100
5 1
2 100
1 1
8 100
dtype: int64
>>> s.idxmax()
3
"""
sdf = self._internal.spark_frame
scol = self.spark.column
index_scols = self._internal.index_spark_columns
# desc_nulls_(last|first) is used via Py4J directly because
# it's not supported in Spark 2.3.
if skipna:
sdf = sdf.orderBy(Column(scol._jc.desc_nulls_last()), NATURAL_ORDER_COLUMN_NAME)
else:
sdf = sdf.orderBy(Column(scol._jc.desc_nulls_first()), NATURAL_ORDER_COLUMN_NAME)
results = sdf.select([scol] + index_scols).take(1)
if len(results) == 0:
raise ValueError("attempt to get idxmin of an empty sequence")
if results[0][0] is None:
# This will only happens when skipna is False because we will
# place nulls first.
return np.nan
values = list(results[0][1:])
if len(values) == 1:
return values[0]
else:
return tuple(values)
def idxmin(self, skipna: bool = True) -> Union[Tuple, Any]:
"""
Return the row label of the minimum value.
If multiple values equal the minimum, the first row label with that
value is returned.
Parameters
----------
skipna : bool, default True
Exclude NA/null values. If the entire Series is NA, the result
will be NA.
Returns
-------
Index
Label of the minimum value.
Raises
------
ValueError
If the Series is empty.
See Also
--------
Series.idxmax : Return index *label* of the first occurrence
of maximum of values.
Notes
-----
This method is the Series version of ``ndarray.argmin``. This method
returns the label of the minimum, while ``ndarray.argmin`` returns
the position. To get the position, use ``series.values.argmin()``.
Examples
--------
>>> s = ps.Series(data=[1, None, 4, 0],
... index=['A', 'B', 'C', 'D'])
>>> s
A 1.0
B NaN
C 4.0
D 0.0
dtype: float64
>>> s.idxmin()
'D'
If `skipna` is False and there is an NA value in the data,
the function returns ``nan``.
>>> s.idxmin(skipna=False)
nan
In case of multi-index, you get a tuple:
>>> index = pd.MultiIndex.from_arrays([
... ['a', 'a', 'b', 'b'], ['c', 'd', 'e', 'f']], names=('first', 'second'))
>>> s = ps.Series(data=[1, None, 4, 0], index=index)
>>> s
first second
a c 1.0
d NaN
b e 4.0
f 0.0
dtype: float64
>>> s.idxmin()
('b', 'f')
If multiple values equal the minimum, the first row label with that
value is returned.
>>> s = ps.Series([1, 100, 1, 100, 1, 100], index=[10, 3, 5, 2, 1, 8])
>>> s
10 1
3 100
5 1
2 100
1 1
8 100
dtype: int64
>>> s.idxmin()
10
"""
sdf = self._internal.spark_frame
scol = self.spark.column
index_scols = self._internal.index_spark_columns
# asc_nulls_(last|first)is used via Py4J directly because
# it's not supported in Spark 2.3.
if skipna:
sdf = sdf.orderBy(Column(scol._jc.asc_nulls_last()), NATURAL_ORDER_COLUMN_NAME)
else:
sdf = sdf.orderBy(Column(scol._jc.asc_nulls_first()), NATURAL_ORDER_COLUMN_NAME)
results = sdf.select([scol] + index_scols).take(1)
if len(results) == 0:
raise ValueError("attempt to get idxmin of an empty sequence")
if results[0][0] is None:
# This will only happens when skipna is False because we will
# place nulls first.
return np.nan
values = list(results[0][1:])
if len(values) == 1:
return values[0]
else:
return tuple(values)
def pop(self, item: Name) -> Union["Series", Scalar]:
"""
Return item and drop from series.
Parameters
----------
item : label
Label of index to be popped.
Returns
-------
Value that is popped from series.
Examples
--------
>>> s = ps.Series(data=np.arange(3), index=['A', 'B', 'C'])
>>> s
A 0
B 1
C 2
dtype: int64
>>> s.pop('A')
0
>>> s
B 1
C 2
dtype: int64
>>> s = ps.Series(data=np.arange(3), index=['A', 'A', 'C'])
>>> s
A 0
A 1
C 2
dtype: int64
>>> s.pop('A')
A 0
A 1
dtype: int64
>>> s
C 2
dtype: int64
Also support for MultiIndex
>>> midx = pd.MultiIndex([['lama', 'cow', 'falcon'],
... ['speed', 'weight', 'length']],
... [[0, 0, 0, 1, 1, 1, 2, 2, 2],
... [0, 1, 2, 0, 1, 2, 0, 1, 2]])
>>> s = ps.Series([45, 200, 1.2, 30, 250, 1.5, 320, 1, 0.3],
... index=midx)
>>> s
lama speed 45.0
weight 200.0
length 1.2
cow speed 30.0
weight 250.0
length 1.5
falcon speed 320.0
weight 1.0
length 0.3
dtype: float64
>>> s.pop('lama')
speed 45.0
weight 200.0
length 1.2
dtype: float64
>>> s
cow speed 30.0
weight 250.0
length 1.5
falcon speed 320.0
weight 1.0
length 0.3
dtype: float64
Also support for MultiIndex with several indexs.
>>> midx = pd.MultiIndex([['a', 'b', 'c'],
... ['lama', 'cow', 'falcon'],
... ['speed', 'weight', 'length']],
... [[0, 0, 0, 0, 0, 0, 1, 1, 1],
... [0, 0, 0, 1, 1, 1, 2, 2, 2],
... [0, 1, 2, 0, 1, 2, 0, 0, 2]]
... )
>>> s = ps.Series([45, 200, 1.2, 30, 250, 1.5, 320, 1, 0.3],
... index=midx)
>>> s
a lama speed 45.0
weight 200.0
length 1.2
cow speed 30.0
weight 250.0
length 1.5
b falcon speed 320.0
speed 1.0
length 0.3
dtype: float64
>>> s.pop(('a', 'lama'))
speed 45.0
weight 200.0
length 1.2
dtype: float64
>>> s
a cow speed 30.0
weight 250.0
length 1.5
b falcon speed 320.0
speed 1.0
length 0.3
dtype: float64
>>> s.pop(('b', 'falcon', 'speed'))
(b, falcon, speed) 320.0
(b, falcon, speed) 1.0
dtype: float64
"""
if not is_name_like_value(item):
raise TypeError("'key' should be string or tuple that contains strings")
if not is_name_like_tuple(item):
item = (item,)
if self._internal.index_level < len(item):
raise KeyError(
"Key length ({}) exceeds index depth ({})".format(
len(item), self._internal.index_level
)
)
internal = self._internal
scols = internal.index_spark_columns[len(item) :] + [self.spark.column]
rows = [internal.spark_columns[level] == index for level, index in enumerate(item)]
sdf = internal.spark_frame.filter(reduce(lambda x, y: x & y, rows)).select(scols)
psdf = self._drop(item)
self._update_anchor(psdf)
if self._internal.index_level == len(item):
# if spark_frame has one column and one data, return data only without frame
pdf = sdf.limit(2).toPandas()
length = len(pdf)
if length == 1:
return pdf[internal.data_spark_column_names[0]].iloc[0]
item_string = name_like_string(item)
sdf = sdf.withColumn(SPARK_DEFAULT_INDEX_NAME, SF.lit(str(item_string)))
internal = InternalFrame(
spark_frame=sdf,
index_spark_columns=[scol_for(sdf, SPARK_DEFAULT_INDEX_NAME)],
column_labels=[self._column_label],
data_fields=[self._internal.data_fields[0]],
)
return first_series(DataFrame(internal))
else:
internal = internal.copy(
spark_frame=sdf,
index_spark_columns=[
scol_for(sdf, col) for col in internal.index_spark_column_names[len(item) :]
],
index_fields=internal.index_fields[len(item) :],
index_names=self._internal.index_names[len(item) :],
data_spark_columns=[scol_for(sdf, internal.data_spark_column_names[0])],
)
return first_series(DataFrame(internal))
def copy(self, deep: bool = True) -> "Series":
"""
Make a copy of this object's indices and data.
Parameters
----------
deep : bool, default True
this parameter is not supported but just dummy parameter to match pandas.
Returns
-------
copy : Series
Examples
--------
>>> s = ps.Series([1, 2], index=["a", "b"])
>>> s
a 1
b 2
dtype: int64
>>> s_copy = s.copy()
>>> s_copy
a 1
b 2
dtype: int64
"""
return first_series(DataFrame(self._internal))
def mode(self, dropna: bool = True) -> "Series":
"""
Return the mode(s) of the dataset.
Always returns Series even if only one value is returned.
Parameters
----------
dropna : bool, default True
Don't consider counts of NaN/NaT.
Returns
-------
Series
Modes of the Series.
Examples
--------
>>> s = ps.Series([0, 0, 1, 1, 1, np.nan, np.nan, np.nan])
>>> s
0 0.0
1 0.0
2 1.0
3 1.0
4 1.0
5 NaN
6 NaN
7 NaN
dtype: float64
>>> s.mode()
0 1.0
dtype: float64
If there are several same modes, all items are shown
>>> s = ps.Series([0, 0, 1, 1, 1, 2, 2, 2, 3, 3, 3,
... np.nan, np.nan, np.nan])
>>> s
0 0.0
1 0.0
2 1.0
3 1.0
4 1.0
5 2.0
6 2.0
7 2.0
8 3.0
9 3.0
10 3.0
11 NaN
12 NaN
13 NaN
dtype: float64
>>> s.mode().sort_values() # doctest: +NORMALIZE_WHITESPACE, +ELLIPSIS
<BLANKLINE>
... 1.0
... 2.0
... 3.0
dtype: float64
With 'dropna' set to 'False', we can also see NaN in the result
>>> s.mode(False).sort_values() # doctest: +NORMALIZE_WHITESPACE, +ELLIPSIS
<BLANKLINE>
... 1.0
... 2.0
... 3.0
... NaN
dtype: float64
"""
ser_count = self.value_counts(dropna=dropna, sort=False)
sdf_count = ser_count._internal.spark_frame
most_value = ser_count.max()
sdf_most_value = sdf_count.filter("count == {}".format(most_value))
sdf = sdf_most_value.select(
F.col(SPARK_DEFAULT_INDEX_NAME).alias(SPARK_DEFAULT_SERIES_NAME)
)
internal = InternalFrame(spark_frame=sdf, index_spark_columns=None, column_labels=[None])
return first_series(DataFrame(internal))
def keys(self) -> "ps.Index":
"""
Return alias for index.
Returns
-------
Index
Index of the Series.
Examples
--------
>>> midx = pd.MultiIndex([['lama', 'cow', 'falcon'],
... ['speed', 'weight', 'length']],
... [[0, 0, 0, 1, 1, 1, 2, 2, 2],
... [0, 1, 2, 0, 1, 2, 0, 1, 2]])
>>> psser = ps.Series([45, 200, 1.2, 30, 250, 1.5, 320, 1, 0.3], index=midx)
>>> psser.keys() # doctest: +SKIP
MultiIndex([( 'lama', 'speed'),
( 'lama', 'weight'),
( 'lama', 'length'),
( 'cow', 'speed'),
( 'cow', 'weight'),
( 'cow', 'length'),
('falcon', 'speed'),
('falcon', 'weight'),
('falcon', 'length')],
)
"""
return self.index
# TODO: 'regex', 'method' parameter
def replace(
self,
to_replace: Optional[Union[Any, List, Tuple, Dict]] = None,
value: Optional[Union[List, Tuple]] = None,
regex: bool = False,
) -> "Series":
"""
Replace values given in to_replace with value.
Values of the Series are replaced with other values dynamically.
Parameters
----------
to_replace : str, list, tuple, dict, Series, int, float, or None
How to find the values that will be replaced.
* numeric, str:
- numeric: numeric values equal to to_replace will be replaced with value
- str: string exactly matching to_replace will be replaced with value
* list of str or numeric:
- if to_replace and value are both lists or tuples, they must be the same length.
- str and numeric rules apply as above.
* dict:
- Dicts can be used to specify different replacement values for different
existing values.
For example, {'a': 'b', 'y': 'z'} replaces the value ‘a’ with ‘b’ and ‘y’
with ‘z’. To use a dict in this way the value parameter should be None.
- For a DataFrame a dict can specify that different values should be replaced
in different columns. For example, {'a': 1, 'b': 'z'} looks for the value 1
in column ‘a’ and the value ‘z’ in column ‘b’ and replaces these values with
whatever is specified in value.
The value parameter should not be None in this case.
You can treat this as a special case of passing two lists except that you are
specifying the column to search in.
See the examples section for examples of each of these.
value : scalar, dict, list, tuple, str default None
Value to replace any values matching to_replace with.
For a DataFrame a dict of values can be used to specify which value to use
for each column (columns not in the dict will not be filled).
Regular expressions, strings and lists or dicts of such objects are also allowed.
Returns
-------
Series
Object after replacement.
Examples
--------
Scalar `to_replace` and `value`
>>> s = ps.Series([0, 1, 2, 3, 4])
>>> s
0 0
1 1
2 2
3 3
4 4
dtype: int64
>>> s.replace(0, 5)
0 5
1 1
2 2
3 3
4 4
dtype: int64
List-like `to_replace`
>>> s.replace([0, 4], 5000)
0 5000
1 1
2 2
3 3
4 5000
dtype: int64
>>> s.replace([1, 2, 3], [10, 20, 30])
0 0
1 10
2 20
3 30
4 4
dtype: int64
Dict-like `to_replace`
>>> s.replace({1: 1000, 2: 2000, 3: 3000, 4: 4000})
0 0
1 1000
2 2000
3 3000
4 4000
dtype: int64
Also support for MultiIndex
>>> midx = pd.MultiIndex([['lama', 'cow', 'falcon'],
... ['speed', 'weight', 'length']],
... [[0, 0, 0, 1, 1, 1, 2, 2, 2],
... [0, 1, 2, 0, 1, 2, 0, 1, 2]])
>>> s = ps.Series([45, 200, 1.2, 30, 250, 1.5, 320, 1, 0.3],
... index=midx)
>>> s
lama speed 45.0
weight 200.0
length 1.2
cow speed 30.0
weight 250.0
length 1.5
falcon speed 320.0
weight 1.0
length 0.3
dtype: float64
>>> s.replace(45, 450)
lama speed 450.0
weight 200.0
length 1.2
cow speed 30.0
weight 250.0
length 1.5
falcon speed 320.0
weight 1.0
length 0.3
dtype: float64
>>> s.replace([45, 30, 320], 500)
lama speed 500.0
weight 200.0
length 1.2
cow speed 500.0
weight 250.0
length 1.5
falcon speed 500.0
weight 1.0
length 0.3
dtype: float64
>>> s.replace({45: 450, 30: 300})
lama speed 450.0
weight 200.0
length 1.2
cow speed 300.0
weight 250.0
length 1.5
falcon speed 320.0
weight 1.0
length 0.3
dtype: float64
"""
if to_replace is None:
return self.fillna(method="ffill")
if not isinstance(to_replace, (str, list, tuple, dict, int, float)):
raise TypeError("'to_replace' should be one of str, list, tuple, dict, int, float")
if regex:
raise NotImplementedError("replace currently not support for regex")
to_replace = list(to_replace) if isinstance(to_replace, tuple) else to_replace
value = list(value) if isinstance(value, tuple) else value
if isinstance(to_replace, list) and isinstance(value, list):
if not len(to_replace) == len(value):
raise ValueError(
"Replacement lists must match in length. Expecting {} got {}".format(
len(to_replace), len(value)
)
)
to_replace = {k: v for k, v in zip(to_replace, value)}
if isinstance(to_replace, dict):
is_start = True
if len(to_replace) == 0:
current = self.spark.column
else:
for to_replace_, value in to_replace.items():
cond = (
(F.isnan(self.spark.column) | self.spark.column.isNull())
if pd.isna(to_replace_)
else (self.spark.column == SF.lit(to_replace_))
)
if is_start:
current = F.when(cond, value)
is_start = False
else:
current = current.when(cond, value)
current = current.otherwise(self.spark.column)
else:
cond = self.spark.column.isin(to_replace)
# to_replace may be a scalar
if np.array(pd.isna(to_replace)).any():
cond = cond | F.isnan(self.spark.column) | self.spark.column.isNull()
current = F.when(cond, value).otherwise(self.spark.column)
return self._with_new_scol(current) # TODO: dtype?
def update(self, other: "Series") -> None:
"""
Modify Series in place using non-NA values from passed Series. Aligns on index.
Parameters
----------
other : Series
Examples
--------
>>> from pyspark.pandas.config import set_option, reset_option
>>> set_option("compute.ops_on_diff_frames", True)
>>> s = ps.Series([1, 2, 3])
>>> s.update(ps.Series([4, 5, 6]))
>>> s.sort_index()
0 4
1 5
2 6
dtype: int64
>>> s = ps.Series(['a', 'b', 'c'])
>>> s.update(ps.Series(['d', 'e'], index=[0, 2]))
>>> s.sort_index()
0 d
1 b
2 e
dtype: object
>>> s = ps.Series([1, 2, 3])
>>> s.update(ps.Series([4, 5, 6, 7, 8]))
>>> s.sort_index()
0 4
1 5
2 6
dtype: int64
>>> s = ps.Series([1, 2, 3], index=[10, 11, 12])
>>> s
10 1
11 2
12 3
dtype: int64
>>> s.update(ps.Series([4, 5, 6]))
>>> s.sort_index()
10 1
11 2
12 3
dtype: int64
>>> s.update(ps.Series([4, 5, 6], index=[11, 12, 13]))
>>> s.sort_index()
10 1
11 4
12 5
dtype: int64
If ``other`` contains NaNs the corresponding values are not updated
in the original Series.
>>> s = ps.Series([1, 2, 3])
>>> s.update(ps.Series([4, np.nan, 6]))
>>> s.sort_index()
0 4.0
1 2.0
2 6.0
dtype: float64
>>> reset_option("compute.ops_on_diff_frames")
"""
if not isinstance(other, Series):
raise TypeError("'other' must be a Series")
combined = combine_frames(self._psdf, other._psdf, how="leftouter")
this_scol = combined["this"]._internal.spark_column_for(self._column_label)
that_scol = combined["that"]._internal.spark_column_for(other._column_label)
scol = (
F.when(that_scol.isNotNull(), that_scol)
.otherwise(this_scol)
.alias(self._psdf._internal.spark_column_name_for(self._column_label))
)
internal = combined["this"]._internal.with_new_spark_column(
self._column_label, scol # TODO: dtype?
)
self._psdf._update_internal_frame(internal.resolved_copy, requires_same_anchor=False)
def where(self, cond: "Series", other: Any = np.nan) -> "Series":
"""
Replace values where the condition is False.
Parameters
----------
cond : boolean Series
Where cond is True, keep the original value. Where False,
replace with corresponding value from other.
other : scalar, Series
Entries where cond is False are replaced with corresponding value from other.
Returns
-------
Series
Examples
--------
>>> from pyspark.pandas.config import set_option, reset_option
>>> set_option("compute.ops_on_diff_frames", True)
>>> s1 = ps.Series([0, 1, 2, 3, 4])
>>> s2 = ps.Series([100, 200, 300, 400, 500])
>>> s1.where(s1 > 0).sort_index()
0 NaN
1 1.0
2 2.0
3 3.0
4 4.0
dtype: float64
>>> s1.where(s1 > 1, 10).sort_index()
0 10
1 10
2 2
3 3
4 4
dtype: int64
>>> s1.where(s1 > 1, s1 + 100).sort_index()
0 100
1 101
2 2
3 3
4 4
dtype: int64
>>> s1.where(s1 > 1, s2).sort_index()
0 100
1 200
2 2
3 3
4 4
dtype: int64
>>> reset_option("compute.ops_on_diff_frames")
"""
assert isinstance(cond, Series)
# We should check the DataFrame from both `cond` and `other`.
should_try_ops_on_diff_frame = not same_anchor(cond, self) or (
isinstance(other, Series) and not same_anchor(other, self)
)
if should_try_ops_on_diff_frame:
# Try to perform it with 'compute.ops_on_diff_frame' option.
psdf = self.to_frame()
tmp_cond_col = verify_temp_column_name(psdf, "__tmp_cond_col__")
tmp_other_col = verify_temp_column_name(psdf, "__tmp_other_col__")
psdf[tmp_cond_col] = cond
psdf[tmp_other_col] = other
# above logic makes a Spark DataFrame looks like below:
# +-----------------+---+----------------+-----------------+
# |__index_level_0__| 0|__tmp_cond_col__|__tmp_other_col__|
# +-----------------+---+----------------+-----------------+
# | 0| 0| false| 100|
# | 1| 1| false| 200|
# | 3| 3| true| 400|
# | 2| 2| true| 300|
# | 4| 4| true| 500|
# +-----------------+---+----------------+-----------------+
condition = (
F.when(
psdf[tmp_cond_col].spark.column,
psdf._psser_for(psdf._internal.column_labels[0]).spark.column,
)
.otherwise(psdf[tmp_other_col].spark.column)
.alias(psdf._internal.data_spark_column_names[0])
)
internal = psdf._internal.with_new_columns(
[condition], column_labels=self._internal.column_labels
)
return first_series(DataFrame(internal))
else:
if isinstance(other, Series):
other = other.spark.column
condition = (
F.when(cond.spark.column, self.spark.column)
.otherwise(other)
.alias(self._internal.data_spark_column_names[0])
)
return self._with_new_scol(condition)
def mask(self, cond: "Series", other: Any = np.nan) -> "Series":
"""
Replace values where the condition is True.
Parameters
----------
cond : boolean Series
Where cond is False, keep the original value. Where True,
replace with corresponding value from other.
other : scalar, Series
Entries where cond is True are replaced with corresponding value from other.
Returns
-------
Series
Examples
--------
>>> from pyspark.pandas.config import set_option, reset_option
>>> set_option("compute.ops_on_diff_frames", True)
>>> s1 = ps.Series([0, 1, 2, 3, 4])
>>> s2 = ps.Series([100, 200, 300, 400, 500])
>>> s1.mask(s1 > 0).sort_index()
0 0.0
1 NaN
2 NaN
3 NaN
4 NaN
dtype: float64
>>> s1.mask(s1 > 1, 10).sort_index()
0 0
1 1
2 10
3 10
4 10
dtype: int64
>>> s1.mask(s1 > 1, s1 + 100).sort_index()
0 0
1 1
2 102
3 103
4 104
dtype: int64
>>> s1.mask(s1 > 1, s2).sort_index()
0 0
1 1
2 300
3 400
4 500
dtype: int64
>>> reset_option("compute.ops_on_diff_frames")
"""
return self.where(cast(Series, ~cond), other)
def xs(self, key: Name, level: Optional[int] = None) -> "Series":
"""
Return cross-section from the Series.
This method takes a `key` argument to select data at a particular
level of a MultiIndex.
Parameters
----------
key : label or tuple of label
Label contained in the index, or partially in a MultiIndex.
level : object, defaults to first n levels (n=1 or len(key))
In case of a key partially contained in a MultiIndex, indicate
which levels are used. Levels can be referred by label or position.
Returns
-------
Series
Cross-section from the original Series
corresponding to the selected index levels.
Examples
--------
>>> midx = pd.MultiIndex([['a', 'b', 'c'],
... ['lama', 'cow', 'falcon'],
... ['speed', 'weight', 'length']],
... [[0, 0, 0, 1, 1, 1, 2, 2, 2],
... [0, 0, 0, 1, 1, 1, 2, 2, 2],
... [0, 1, 2, 0, 1, 2, 0, 1, 2]])
>>> s = ps.Series([45, 200, 1.2, 30, 250, 1.5, 320, 1, 0.3],
... index=midx)
>>> s
a lama speed 45.0
weight 200.0
length 1.2
b cow speed 30.0
weight 250.0
length 1.5
c falcon speed 320.0
weight 1.0
length 0.3
dtype: float64
Get values at specified index
>>> s.xs('a')
lama speed 45.0
weight 200.0
length 1.2
dtype: float64
Get values at several indexes
>>> s.xs(('a', 'lama'))
speed 45.0
weight 200.0
length 1.2
dtype: float64
Get values at specified index and level
>>> s.xs('lama', level=1)
a speed 45.0
weight 200.0
length 1.2
dtype: float64
"""
if not isinstance(key, tuple):
key = (key,)
if level is None:
level = 0
internal = self._internal
scols = (
internal.index_spark_columns[:level]
+ internal.index_spark_columns[level + len(key) :]
+ [self.spark.column]
)
rows = [internal.spark_columns[lvl] == index for lvl, index in enumerate(key, level)]
sdf = internal.spark_frame.filter(reduce(lambda x, y: x & y, rows)).select(scols)
if internal.index_level == len(key):
# if spark_frame has one column and one data, return data only without frame
pdf = sdf.limit(2).toPandas()
length = len(pdf)
if length == 1:
return pdf[self._internal.data_spark_column_names[0]].iloc[0]
index_spark_column_names = (
internal.index_spark_column_names[:level]
+ internal.index_spark_column_names[level + len(key) :]
)
index_names = internal.index_names[:level] + internal.index_names[level + len(key) :]
index_fields = internal.index_fields[:level] + internal.index_fields[level + len(key) :]
internal = internal.copy(
spark_frame=sdf,
index_spark_columns=[scol_for(sdf, col) for col in index_spark_column_names],
index_names=index_names,
index_fields=index_fields,
data_spark_columns=[scol_for(sdf, internal.data_spark_column_names[0])],
)
return first_series(DataFrame(internal))
def pct_change(self, periods: int = 1) -> "Series":
"""
Percentage change between the current and a prior element.
.. note:: the current implementation of this API uses Spark's Window without
specifying partition specification. This leads to move all data into
single partition in single machine and could cause serious
performance degradation. Avoid this method against very large dataset.
Parameters
----------
periods : int, default 1
Periods to shift for forming percent change.
Returns
-------
Series
Examples
--------
>>> psser = ps.Series([90, 91, 85], index=[2, 4, 1])
>>> psser
2 90
4 91
1 85
dtype: int64
>>> psser.pct_change()
2 NaN
4 0.011111
1 -0.065934
dtype: float64
>>> psser.sort_index().pct_change()
1 NaN
2 0.058824
4 0.011111
dtype: float64
>>> psser.pct_change(periods=2)
2 NaN
4 NaN
1 -0.055556
dtype: float64
"""
scol = self.spark.column
window = Window.orderBy(NATURAL_ORDER_COLUMN_NAME).rowsBetween(-periods, -periods)
prev_row = F.lag(scol, periods).over(window)
return self._with_new_scol((scol - prev_row) / prev_row).spark.analyzed
def combine_first(self, other: "Series") -> "Series":
"""
Combine Series values, choosing the calling Series's values first.
Parameters
----------
other : Series
The value(s) to be combined with the `Series`.
Returns
-------
Series
The result of combining the Series with the other object.
See Also
--------
Series.combine : Perform elementwise operation on two Series
using a given function.
Notes
-----
Result index will be the union of the two indexes.
Examples
--------
>>> s1 = ps.Series([1, np.nan])
>>> s2 = ps.Series([3, 4])
>>> with ps.option_context("compute.ops_on_diff_frames", True):
... s1.combine_first(s2)
0 1.0
1 4.0
dtype: float64
"""
if not isinstance(other, ps.Series):
raise TypeError("`combine_first` only allows `Series` for parameter `other`")
if same_anchor(self, other):
this = self.spark.column
that = other.spark.column
combined = self._psdf
else:
combined = combine_frames(self._psdf, other._psdf)
this = combined["this"]._internal.spark_column_for(self._column_label)
that = combined["that"]._internal.spark_column_for(other._column_label)
# If `self` has missing value, use value of `other`
cond = F.when(this.isNull(), that).otherwise(this)
# If `self` and `other` come from same frame, the anchor should be kept
if same_anchor(self, other):
return self._with_new_scol(cond) # TODO: dtype?
index_scols = combined._internal.index_spark_columns
sdf = combined._internal.spark_frame.select(
*index_scols, cond.alias(self._internal.data_spark_column_names[0])
).distinct()
internal = self._internal.with_new_sdf(
sdf, index_fields=combined._internal.index_fields, data_fields=[None] # TODO: dtype?
)
return first_series(DataFrame(internal))
def dot(self, other: Union["Series", DataFrame]) -> Union[Scalar, "Series"]:
"""
Compute the dot product between the Series and the columns of other.
This method computes the dot product between the Series and another
one, or the Series and each columns of a DataFrame.
It can also be called using `self @ other` in Python >= 3.5.
.. note:: This API is slightly different from pandas when indexes from both Series
are not aligned. To match with pandas', it requires to read the whole data for,
for example, counting. pandas raises an exception; however, pandas-on-Spark
just proceeds and performs by ignoring mismatches with NaN permissively.
>>> pdf1 = pd.Series([1, 2, 3], index=[0, 1, 2])
>>> pdf2 = pd.Series([1, 2, 3], index=[0, 1, 3])
>>> pdf1.dot(pdf2) # doctest: +SKIP
...
ValueError: matrices are not aligned
>>> psdf1 = ps.Series([1, 2, 3], index=[0, 1, 2])
>>> psdf2 = ps.Series([1, 2, 3], index=[0, 1, 3])
>>> psdf1.dot(psdf2) # doctest: +SKIP
5
Parameters
----------
other : Series, DataFrame.
The other object to compute the dot product with its columns.
Returns
-------
scalar, Series
Return the dot product of the Series and other if other is a
Series, the Series of the dot product of Series and each rows of
other if other is a DataFrame.
Notes
-----
The Series and other has to share the same index if other is a Series
or a DataFrame.
Examples
--------
>>> s = ps.Series([0, 1, 2, 3])
>>> s.dot(s)
14
>>> s @ s
14
>>> psdf = ps.DataFrame({'x': [0, 1, 2, 3], 'y': [0, -1, -2, -3]})
>>> psdf
x y
0 0 0
1 1 -1
2 2 -2
3 3 -3
>>> with ps.option_context("compute.ops_on_diff_frames", True):
... s.dot(psdf)
...
x 14
y -14
dtype: int64
"""
if isinstance(other, DataFrame):
if not same_anchor(self, other):
if not self.index.sort_values().equals(other.index.sort_values()):
raise ValueError("matrices are not aligned")
other_copy = other.copy() # type: DataFrame
column_labels = other_copy._internal.column_labels
self_column_label = verify_temp_column_name(other_copy, "__self_column__")
other_copy[self_column_label] = self
self_psser = other_copy._psser_for(self_column_label)
product_pssers = [
cast(Series, other_copy._psser_for(label) * self_psser) for label in column_labels
]
dot_product_psser = DataFrame(
other_copy._internal.with_new_columns(product_pssers, column_labels=column_labels)
).sum()
return cast(Series, dot_product_psser).rename(self.name)
else:
assert isinstance(other, Series)
if not same_anchor(self, other):
if len(self.index) != len(other.index):
raise ValueError("matrices are not aligned")
return (self * other).sum()
def __matmul__(self, other: Union["Series", DataFrame]) -> Union[Scalar, "Series"]:
"""
Matrix multiplication using binary `@` operator in Python>=3.5.
"""
return self.dot(other)
def repeat(self, repeats: Union[int, "Series"]) -> "Series":
"""
Repeat elements of a Series.
Returns a new Series where each element of the current Series
is repeated consecutively a given number of times.
Parameters
----------
repeats : int or Series
The number of repetitions for each element. This should be a
non-negative integer. Repeating 0 times will return an empty
Series.
Returns
-------
Series
Newly created Series with repeated elements.
See Also
--------
Index.repeat : Equivalent function for Index.
Examples
--------
>>> s = ps.Series(['a', 'b', 'c'])
>>> s
0 a
1 b
2 c
dtype: object
>>> s.repeat(2)
0 a
1 b
2 c
0 a
1 b
2 c
dtype: object
>>> ps.Series([1, 2, 3]).repeat(0)
Series([], dtype: int64)
"""
if not isinstance(repeats, (int, Series)):
raise TypeError(
"`repeats` argument must be integer or Series, but got {}".format(type(repeats))
)
if isinstance(repeats, Series):
if not same_anchor(self, repeats):
psdf = self.to_frame()
temp_repeats = verify_temp_column_name(psdf, "__temp_repeats__")
psdf[temp_repeats] = repeats
return (
psdf._psser_for(psdf._internal.column_labels[0])
.repeat(psdf[temp_repeats])
.rename(self.name)
)
else:
scol = F.explode(
F.array_repeat(self.spark.column, repeats.astype("int32").spark.column)
).alias(name_like_string(self.name))
sdf = self._internal.spark_frame.select(self._internal.index_spark_columns + [scol])
internal = self._internal.copy(
spark_frame=sdf,
index_spark_columns=[
scol_for(sdf, col) for col in self._internal.index_spark_column_names
],
data_spark_columns=[scol_for(sdf, name_like_string(self.name))],
)
return first_series(DataFrame(internal))
else:
if repeats < 0:
raise ValueError("negative dimensions are not allowed")
psdf = self._psdf[[self.name]]
if repeats == 0:
return first_series(DataFrame(psdf._internal.with_filter(SF.lit(False))))
else:
return first_series(ps.concat([psdf] * repeats))
def asof(self, where: Union[Any, List]) -> Union[Scalar, "Series"]:
"""
Return the last row(s) without any NaNs before `where`.
The last row (for each element in `where`, if list) without any
NaN is taken.
If there is no good value, NaN is returned.
.. note:: This API is dependent on :meth:`Index.is_monotonic_increasing`
which can be expensive.
Parameters
----------
where : index or array-like of indices
Returns
-------
scalar or Series
The return can be:
* scalar : when `self` is a Series and `where` is a scalar
* Series: when `self` is a Series and `where` is an array-like
Return scalar or Series
Notes
-----
Indices are assumed to be sorted. Raises if this is not the case.
Examples
--------
>>> s = ps.Series([1, 2, np.nan, 4], index=[10, 20, 30, 40])
>>> s
10 1.0
20 2.0
30 NaN
40 4.0
dtype: float64
A scalar `where`.
>>> s.asof(20)
2.0
For a sequence `where`, a Series is returned. The first value is
NaN, because the first element of `where` is before the first
index value.
>>> s.asof([5, 20]).sort_index()
5 NaN
20 2.0
dtype: float64
Missing values are not considered. The following is ``2.0``, not
NaN, even though NaN is at the index location for ``30``.
>>> s.asof(30)
2.0
"""
should_return_series = True
if isinstance(self.index, ps.MultiIndex):
raise ValueError("asof is not supported for a MultiIndex")
if isinstance(where, (ps.Index, ps.Series, DataFrame)):
raise ValueError("where cannot be an Index, Series or a DataFrame")
if not self.index.is_monotonic_increasing:
raise ValueError("asof requires a sorted index")
if not is_list_like(where):
should_return_series = False
where = [where]
index_scol = self._internal.index_spark_columns[0]
index_type = self._internal.spark_type_for(index_scol)
cond = [
F.max(F.when(index_scol <= SF.lit(index).cast(index_type), self.spark.column))
for index in where
]
sdf = self._internal.spark_frame.select(cond)
if not should_return_series:
with sql_conf({SPARK_CONF_ARROW_ENABLED: False}):
# Disable Arrow to keep row ordering.
result = cast(pd.DataFrame, sdf.limit(1).toPandas()).iloc[0, 0]
return result if result is not None else np.nan
# The data is expected to be small so it's fine to transpose/use default index.
with ps.option_context("compute.default_index_type", "distributed", "compute.max_rows", 1):
psdf = ps.DataFrame(sdf) # type: DataFrame
psdf.columns = pd.Index(where)
return first_series(psdf.transpose()).rename(self.name)
def mad(self) -> float:
"""
Return the mean absolute deviation of values.
Examples
--------
>>> s = ps.Series([1, 2, 3, 4])
>>> s
0 1
1 2
2 3
3 4
dtype: int64
>>> s.mad()
1.0
"""
sdf = self._internal.spark_frame
spark_column = self.spark.column
avg = unpack_scalar(sdf.select(F.avg(spark_column)))
mad = unpack_scalar(sdf.select(F.avg(F.abs(spark_column - avg))))
return mad
def unstack(self, level: int = -1) -> DataFrame:
"""
Unstack, a.k.a. pivot, Series with MultiIndex to produce DataFrame.
The level involved will automatically get sorted.
Notes
-----
Unlike pandas, pandas-on-Spark doesn't check whether an index is duplicated or not
because the checking of duplicated index requires scanning whole data which
can be quite expensive.
Parameters
----------
level : int, str, or list of these, default last level
Level(s) to unstack, can pass level name.
Returns
-------
DataFrame
Unstacked Series.
Examples
--------
>>> s = ps.Series([1, 2, 3, 4],
... index=pd.MultiIndex.from_product([['one', 'two'],
... ['a', 'b']]))
>>> s
one a 1
b 2
two a 3
b 4
dtype: int64
>>> s.unstack(level=-1).sort_index()
a b
one 1 2
two 3 4
>>> s.unstack(level=0).sort_index()
one two
a 1 3
b 2 4
"""
if not isinstance(self.index, ps.MultiIndex):
raise ValueError("Series.unstack only support for a MultiIndex")
index_nlevels = self.index.nlevels
if level > 0 and (level > index_nlevels - 1):
raise IndexError(
"Too many levels: Index has only {} levels, not {}".format(index_nlevels, level + 1)
)
elif level < 0 and (level < -index_nlevels):
raise IndexError(
"Too many levels: Index has only {} levels, {} is not a valid level number".format(
index_nlevels, level
)
)
internal = self._internal.resolved_copy
index_map = list(
zip(internal.index_spark_column_names, internal.index_names, internal.index_fields)
)
pivot_col, column_label_names, _ = index_map.pop(level)
index_scol_names, index_names, index_fields = zip(*index_map)
col = internal.data_spark_column_names[0]
sdf = internal.spark_frame
sdf = sdf.groupby(list(index_scol_names)).pivot(pivot_col).agg(F.first(scol_for(sdf, col)))
internal = InternalFrame(
spark_frame=sdf,
index_spark_columns=[scol_for(sdf, col) for col in index_scol_names],
index_names=list(index_names),
index_fields=list(index_fields),
column_label_names=[column_label_names],
)
internal = internal.copy(
data_fields=[
field.copy(dtype=self._internal.data_fields[0].dtype)
for field in internal.data_fields
]
)
return DataFrame(internal)
def item(self) -> Scalar:
"""
Return the first element of the underlying data as a Python scalar.
Returns
-------
scalar
The first element of Series.
Raises
------
ValueError
If the data is not length-1.
Examples
--------
>>> psser = ps.Series([10])
>>> psser.item()
10
"""
return self.head(2)._to_internal_pandas().item()
def iteritems(self) -> Iterable[Tuple[Name, Any]]:
"""
Lazily iterate over (index, value) tuples.
This method returns an iterable tuple (index, value). This is
convenient if you want to create a lazy iterator.
.. note:: Unlike pandas', the iteritems in pandas-on-Spark returns generator rather
zip object
Returns
-------
iterable
Iterable of tuples containing the (index, value) pairs from a
Series.
See Also
--------
DataFrame.items : Iterate over (column name, Series) pairs.
DataFrame.iterrows : Iterate over DataFrame rows as (index, Series) pairs.
Examples
--------
>>> s = ps.Series(['A', 'B', 'C'])
>>> for index, value in s.items():
... print("Index : {}, Value : {}".format(index, value))
Index : 0, Value : A
Index : 1, Value : B
Index : 2, Value : C
"""
internal_index_columns = self._internal.index_spark_column_names
internal_data_column = self._internal.data_spark_column_names[0]
def extract_kv_from_spark_row(row: Row) -> Tuple[Name, Any]:
k = (
row[internal_index_columns[0]]
if len(internal_index_columns) == 1
else tuple(row[c] for c in internal_index_columns)
)
v = row[internal_data_column]
return k, v
for k, v in map(
extract_kv_from_spark_row, self._internal.resolved_copy.spark_frame.toLocalIterator()
):
yield k, v
def items(self) -> Iterable[Tuple[Name, Any]]:
"""This is an alias of ``iteritems``."""
return self.iteritems()
def droplevel(self, level: Union[int, Name, List[Union[int, Name]]]) -> "Series":
"""
Return Series with requested index level(s) removed.
Parameters
----------
level : int, str, or list-like
If a string is given, must be the name of a level
If list-like, elements must be names or positional indexes
of levels.
Returns
-------
Series
Series with requested index level(s) removed.
Examples
--------
>>> psser = ps.Series(
... [1, 2, 3],
... index=pd.MultiIndex.from_tuples(
... [("x", "a"), ("x", "b"), ("y", "c")], names=["level_1", "level_2"]
... ),
... )
>>> psser
level_1 level_2
x a 1
b 2
y c 3
dtype: int64
Removing specific index level by level
>>> psser.droplevel(0)
level_2
a 1
b 2
c 3
dtype: int64
Removing specific index level by name
>>> psser.droplevel("level_2")
level_1
x 1
x 2
y 3
dtype: int64
"""
return first_series(self.to_frame().droplevel(level=level, axis=0)).rename(self.name)
def tail(self, n: int = 5) -> "Series":
"""
Return the last `n` rows.
This function returns last `n` rows from the object based on
position. It is useful for quickly verifying data, for example,
after sorting or appending rows.
For negative values of `n`, this function returns all rows except
the first `n` rows, equivalent to ``df[n:]``.
Parameters
----------
n : int, default 5
Number of rows to select.
Returns
-------
type of caller
The last `n` rows of the caller object.
See Also
--------
DataFrame.head : The first `n` rows of the caller object.
Examples
--------
>>> psser = ps.Series([1, 2, 3, 4, 5])
>>> psser
0 1
1 2
2 3
3 4
4 5
dtype: int64
>>> psser.tail(3) # doctest: +SKIP
2 3
3 4
4 5
dtype: int64
"""
return first_series(self.to_frame().tail(n=n)).rename(self.name)
def explode(self) -> "Series":
"""
Transform each element of a list-like to a row.
Returns
-------
Series
Exploded lists to rows; index will be duplicated for these rows.
See Also
--------
Series.str.split : Split string values on specified separator.
Series.unstack : Unstack, a.k.a. pivot, Series with MultiIndex
to produce DataFrame.
DataFrame.melt : Unpivot a DataFrame from wide format to long format.
DataFrame.explode : Explode a DataFrame from list-like
columns to long format.
Examples
--------
>>> psser = ps.Series([[1, 2, 3], [], [3, 4]])
>>> psser
0 [1, 2, 3]
1 []
2 [3, 4]
dtype: object
>>> psser.explode() # doctest: +SKIP
0 1.0
0 2.0
0 3.0
1 NaN
2 3.0
2 4.0
dtype: float64
"""
if not isinstance(self.spark.data_type, ArrayType):
return self.copy()
scol = F.explode_outer(self.spark.column).alias(name_like_string(self._column_label))
internal = self._internal.with_new_columns([scol], keep_order=False)
return first_series(DataFrame(internal))
def argsort(self) -> "Series":
"""
Return the integer indices that would sort the Series values.
Unlike pandas, the index order is not preserved in the result.
Returns
-------
Series
Positions of values within the sort order with -1 indicating
nan values.
Examples
--------
>>> psser = ps.Series([3, 3, 4, 1, 6, 2, 3, 7, 8, 7, 10])
>>> psser
0 3
1 3
2 4
3 1
4 6
5 2
6 3
7 7
8 8
9 7
10 10
dtype: int64
>>> psser.argsort().sort_index()
0 3
1 5
2 0
3 1
4 6
5 2
6 4
7 7
8 9
9 8
10 10
dtype: int64
"""
notnull = self.loc[self.notnull()]
sdf_for_index = notnull._internal.spark_frame.select(notnull._internal.index_spark_columns)
tmp_join_key = verify_temp_column_name(sdf_for_index, "__tmp_join_key__")
sdf_for_index, _ = InternalFrame.attach_distributed_sequence_column(
sdf_for_index, tmp_join_key
)
# sdf_for_index:
# +----------------+-----------------+
# |__tmp_join_key__|__index_level_0__|
# +----------------+-----------------+
# | 0| 0|
# | 1| 1|
# | 2| 2|
# | 3| 3|
# | 4| 4|
# +----------------+-----------------+
sdf_for_data = notnull._internal.spark_frame.select(
notnull.spark.column.alias("values"), NATURAL_ORDER_COLUMN_NAME
)
sdf_for_data, _ = InternalFrame.attach_distributed_sequence_column(
sdf_for_data, SPARK_DEFAULT_SERIES_NAME
)
# sdf_for_data:
# +---+------+-----------------+
# | 0|values|__natural_order__|
# +---+------+-----------------+
# | 0| 3| 25769803776|
# | 1| 3| 51539607552|
# | 2| 4| 77309411328|
# | 3| 1| 103079215104|
# | 4| 2| 128849018880|
# +---+------+-----------------+
sdf_for_data = sdf_for_data.sort(
scol_for(sdf_for_data, "values"), NATURAL_ORDER_COLUMN_NAME
).drop("values", NATURAL_ORDER_COLUMN_NAME)
tmp_join_key = verify_temp_column_name(sdf_for_data, "__tmp_join_key__")
sdf_for_data, _ = InternalFrame.attach_distributed_sequence_column(
sdf_for_data, tmp_join_key
)
# sdf_for_index: sdf_for_data:
# +----------------+-----------------+ +----------------+---+
# |__tmp_join_key__|__index_level_0__| |__tmp_join_key__| 0|
# +----------------+-----------------+ +----------------+---+
# | 0| 0| | 0| 3|
# | 1| 1| | 1| 4|
# | 2| 2| | 2| 0|
# | 3| 3| | 3| 1|
# | 4| 4| | 4| 2|
# +----------------+-----------------+ +----------------+---+
sdf = sdf_for_index.join(sdf_for_data, on=tmp_join_key).drop(tmp_join_key)
internal = self._internal.with_new_sdf(
spark_frame=sdf,
data_columns=[SPARK_DEFAULT_SERIES_NAME],
index_fields=[
InternalField(dtype=field.dtype) for field in self._internal.index_fields
],
data_fields=[None],
)
psser = first_series(DataFrame(internal))
return cast(
Series,
ps.concat([psser, self.loc[self.isnull()].spark.transform(lambda _: SF.lit(-1))]),
)
def argmax(self) -> int:
"""
Return int position of the largest value in the Series.
If the maximum is achieved in multiple locations,
the first row position is returned.
Returns
-------
int
Row position of the maximum value.
Examples
--------
Consider dataset containing cereal calories
>>> s = ps.Series({'Corn Flakes': 100.0, 'Almond Delight': 110.0,
... 'Cinnamon Toast Crunch': 120.0, 'Cocoa Puff': 110.0})
>>> s # doctest: +SKIP
Corn Flakes 100.0
Almond Delight 110.0
Cinnamon Toast Crunch 120.0
Cocoa Puff 110.0
dtype: float64
>>> s.argmax() # doctest: +SKIP
2
"""
sdf = self._internal.spark_frame.select(self.spark.column, NATURAL_ORDER_COLUMN_NAME)
max_value = sdf.select(
F.max(scol_for(sdf, self._internal.data_spark_column_names[0])),
F.first(NATURAL_ORDER_COLUMN_NAME),
).head()
if max_value[1] is None:
raise ValueError("attempt to get argmax of an empty sequence")
elif max_value[0] is None:
return -1
# We should remember the natural sequence started from 0
seq_col_name = verify_temp_column_name(sdf, "__distributed_sequence_column__")
sdf, _ = InternalFrame.attach_distributed_sequence_column(
sdf.drop(NATURAL_ORDER_COLUMN_NAME), seq_col_name
)
# If the maximum is achieved in multiple locations, the first row position is returned.
return sdf.filter(
scol_for(sdf, self._internal.data_spark_column_names[0]) == max_value[0]
).head()[0]
def argmin(self) -> int:
"""
Return int position of the smallest value in the Series.
If the minimum is achieved in multiple locations,
the first row position is returned.
Returns
-------
int
Row position of the minimum value.
Examples
--------
Consider dataset containing cereal calories
>>> s = ps.Series({'Corn Flakes': 100.0, 'Almond Delight': 110.0,
... 'Cinnamon Toast Crunch': 120.0, 'Cocoa Puff': 110.0})
>>> s # doctest: +SKIP
Corn Flakes 100.0
Almond Delight 110.0
Cinnamon Toast Crunch 120.0
Cocoa Puff 110.0
dtype: float64
>>> s.argmin() # doctest: +SKIP
0
"""
sdf = self._internal.spark_frame.select(self.spark.column, NATURAL_ORDER_COLUMN_NAME)
min_value = sdf.select(
F.min(scol_for(sdf, self._internal.data_spark_column_names[0])),
F.first(NATURAL_ORDER_COLUMN_NAME),
).head()
if min_value[1] is None:
raise ValueError("attempt to get argmin of an empty sequence")
elif min_value[0] is None:
return -1
# We should remember the natural sequence started from 0
seq_col_name = verify_temp_column_name(sdf, "__distributed_sequence_column__")
sdf, _ = InternalFrame.attach_distributed_sequence_column(
sdf.drop(NATURAL_ORDER_COLUMN_NAME), seq_col_name
)
# If the minimum is achieved in multiple locations, the first row position is returned.
return sdf.filter(
scol_for(sdf, self._internal.data_spark_column_names[0]) == min_value[0]
).head()[0]
def compare(
self, other: "Series", keep_shape: bool = False, keep_equal: bool = False
) -> DataFrame:
"""
Compare to another Series and show the differences.
Parameters
----------
other : Series
Object to compare with.
keep_shape : bool, default False
If true, all rows and columns are kept.
Otherwise, only the ones with different values are kept.
keep_equal : bool, default False
If true, the result keeps values that are equal.
Otherwise, equal values are shown as NaNs.
Returns
-------
DataFrame
Notes
-----
Matching NaNs will not appear as a difference.
Examples
--------
>>> from pyspark.pandas.config import set_option, reset_option
>>> set_option("compute.ops_on_diff_frames", True)
>>> s1 = ps.Series(["a", "b", "c", "d", "e"])
>>> s2 = ps.Series(["a", "a", "c", "b", "e"])
Align the differences on columns
>>> s1.compare(s2).sort_index()
self other
1 b a
3 d b
Keep all original rows
>>> s1.compare(s2, keep_shape=True).sort_index()
self other
0 None None
1 b a
2 None None
3 d b
4 None None
Keep all original rows and also all original values
>>> s1.compare(s2, keep_shape=True, keep_equal=True).sort_index()
self other
0 a a
1 b a
2 c c
3 d b
4 e e
>>> reset_option("compute.ops_on_diff_frames")
"""
if same_anchor(self, other):
self_column_label = verify_temp_column_name(other.to_frame(), "__self_column__")
other_column_label = verify_temp_column_name(self.to_frame(), "__other_column__")
combined = DataFrame(
self._internal.with_new_columns(
[self.rename(self_column_label), other.rename(other_column_label)]
)
) # type: DataFrame
else:
if not self.index.equals(other.index):
raise ValueError("Can only compare identically-labeled Series objects")
combined = combine_frames(self.to_frame(), other.to_frame())
this_column_label = "self"
that_column_label = "other"
if keep_equal and keep_shape:
combined.columns = pd.Index([this_column_label, that_column_label])
return combined
this_data_scol = combined._internal.data_spark_columns[0]
that_data_scol = combined._internal.data_spark_columns[1]
index_scols = combined._internal.index_spark_columns
sdf = combined._internal.spark_frame
if keep_shape:
this_scol = (
F.when(this_data_scol == that_data_scol, None)
.otherwise(this_data_scol)
.alias(this_column_label)
)
this_field = combined._internal.data_fields[0].copy(
name=this_column_label, nullable=True
)
that_scol = (
F.when(this_data_scol == that_data_scol, None)
.otherwise(that_data_scol)
.alias(that_column_label)
)
that_field = combined._internal.data_fields[1].copy(
name=that_column_label, nullable=True
)
else:
sdf = sdf.filter(~this_data_scol.eqNullSafe(that_data_scol))
this_scol = this_data_scol.alias(this_column_label)
this_field = combined._internal.data_fields[0].copy(name=this_column_label)
that_scol = that_data_scol.alias(that_column_label)
that_field = combined._internal.data_fields[1].copy(name=that_column_label)
sdf = sdf.select(*index_scols, this_scol, that_scol, NATURAL_ORDER_COLUMN_NAME)
internal = InternalFrame(
spark_frame=sdf,
index_spark_columns=[
scol_for(sdf, col) for col in self._internal.index_spark_column_names
],
index_names=self._internal.index_names,
index_fields=combined._internal.index_fields,
column_labels=[(this_column_label,), (that_column_label,)],
data_spark_columns=[scol_for(sdf, this_column_label), scol_for(sdf, that_column_label)],
data_fields=[this_field, that_field],
column_label_names=[None],
)
return DataFrame(internal)
def align(
self,
other: Union[DataFrame, "Series"],
join: str = "outer",
axis: Optional[Axis] = None,
copy: bool = True,
) -> Tuple["Series", Union[DataFrame, "Series"]]:
"""
Align two objects on their axes with the specified join method.
Join method is specified for each axis Index.
Parameters
----------
other : DataFrame or Series
join : {{'outer', 'inner', 'left', 'right'}}, default 'outer'
axis : allowed axis of the other object, default None
Align on index (0), columns (1), or both (None).
copy : bool, default True
Always returns new objects. If copy=False and no reindexing is
required then original objects are returned.
Returns
-------
(left, right) : (Series, type of other)
Aligned objects.
Examples
--------
>>> ps.set_option("compute.ops_on_diff_frames", True)
>>> s1 = ps.Series([7, 8, 9], index=[10, 11, 12])
>>> s2 = ps.Series(["g", "h", "i"], index=[10, 20, 30])
>>> aligned_l, aligned_r = s1.align(s2)
>>> aligned_l.sort_index()
10 7.0
11 8.0
12 9.0
20 NaN
30 NaN
dtype: float64
>>> aligned_r.sort_index()
10 g
11 None
12 None
20 h
30 i
dtype: object
Align with the join type "inner":
>>> aligned_l, aligned_r = s1.align(s2, join="inner")
>>> aligned_l.sort_index()
10 7
dtype: int64
>>> aligned_r.sort_index()
10 g
dtype: object
Align with a DataFrame:
>>> df = ps.DataFrame({"a": [1, 2, 3], "b": ["a", "b", "c"]}, index=[10, 20, 30])
>>> aligned_l, aligned_r = s1.align(df)
>>> aligned_l.sort_index()
10 7.0
11 8.0
12 9.0
20 NaN
30 NaN
dtype: float64
>>> aligned_r.sort_index()
a b
10 1.0 a
11 NaN None
12 NaN None
20 2.0 b
30 3.0 c
>>> ps.reset_option("compute.ops_on_diff_frames")
"""
axis = validate_axis(axis)
if axis == 1:
raise ValueError("Series does not support columns axis.")
self_df = self.to_frame()
left, right = self_df.align(other, join=join, axis=axis, copy=False)
if left is self_df:
left_ser = self
else:
left_ser = first_series(left).rename(self.name)
return (left_ser.copy(), right.copy()) if copy else (left_ser, right)
def between_time(
self,
start_time: Union[datetime.time, str],
end_time: Union[datetime.time, str],
include_start: bool = True,
include_end: bool = True,
axis: Axis = 0,
) -> "Series":
"""
Select values between particular times of the day (example: 9:00-9:30 AM).
By setting ``start_time`` to be later than ``end_time``,
you can get the times that are *not* between the two times.
Parameters
----------
start_time : datetime.time or str
Initial time as a time filter limit.
end_time : datetime.time or str
End time as a time filter limit.
include_start : bool, default True
Whether the start time needs to be included in the result.
include_end : bool, default True
Whether the end time needs to be included in the result.
axis : {0 or 'index', 1 or 'columns'}, default 0
Determine range time on index or columns value.
Returns
-------
Series
Data from the original object filtered to the specified dates range.
Raises
------
TypeError
If the index is not a :class:`DatetimeIndex`
See Also
--------
at_time : Select values at a particular time of the day.
last : Select final periods of time series based on a date offset.
DatetimeIndex.indexer_between_time : Get just the index locations for
values between particular times of the day.
Examples
--------
>>> idx = pd.date_range('2018-04-09', periods=4, freq='1D20min')
>>> psser = ps.Series([1, 2, 3, 4], index=idx)
>>> psser
2018-04-09 00:00:00 1
2018-04-10 00:20:00 2
2018-04-11 00:40:00 3
2018-04-12 01:00:00 4
dtype: int64
>>> psser.between_time('0:15', '0:45')
2018-04-10 00:20:00 2
2018-04-11 00:40:00 3
dtype: int64
"""
return first_series(
self.to_frame().between_time(start_time, end_time, include_start, include_end, axis)
).rename(self.name)
def at_time(
self, time: Union[datetime.time, str], asof: bool = False, axis: Axis = 0
) -> "Series":
"""
Select values at particular time of day (example: 9:30AM).
Parameters
----------
time : datetime.time or str
axis : {0 or 'index', 1 or 'columns'}, default 0
Returns
-------
Series
Raises
------
TypeError
If the index is not a :class:`DatetimeIndex`
See Also
--------
between_time : Select values between particular times of the day.
DatetimeIndex.indexer_at_time : Get just the index locations for
values at particular time of the day.
Examples
--------
>>> idx = pd.date_range('2018-04-09', periods=4, freq='12H')
>>> psser = ps.Series([1, 2, 3, 4], index=idx)
>>> psser
2018-04-09 00:00:00 1
2018-04-09 12:00:00 2
2018-04-10 00:00:00 3
2018-04-10 12:00:00 4
dtype: int64
>>> psser.at_time('12:00')
2018-04-09 12:00:00 2
2018-04-10 12:00:00 4
dtype: int64
"""
return first_series(self.to_frame().at_time(time, asof, axis)).rename(self.name)
def _cum(
self,
func: Callable[[Column], Column],
skipna: bool,
part_cols: Sequence["ColumnOrName"] = (),
ascending: bool = True,
) -> "Series":
# This is used to cummin, cummax, cumsum, etc.
if ascending:
window = (
Window.orderBy(F.asc(NATURAL_ORDER_COLUMN_NAME))
.partitionBy(*part_cols)
.rowsBetween(Window.unboundedPreceding, Window.currentRow)
)
else:
window = (
Window.orderBy(F.desc(NATURAL_ORDER_COLUMN_NAME))
.partitionBy(*part_cols)
.rowsBetween(Window.unboundedPreceding, Window.currentRow)
)
if skipna:
# There is a behavior difference between pandas and PySpark. In case of cummax,
#
# Input:
# A B
# 0 2.0 1.0
# 1 5.0 NaN
# 2 1.0 0.0
# 3 2.0 4.0
# 4 4.0 9.0
#
# pandas:
# A B
# 0 2.0 1.0
# 1 5.0 NaN
# 2 5.0 1.0
# 3 5.0 4.0
# 4 5.0 9.0
#
# PySpark:
# A B
# 0 2.0 1.0
# 1 5.0 1.0
# 2 5.0 1.0
# 3 5.0 4.0
# 4 5.0 9.0
scol = F.when(
# Manually sets nulls given the column defined above.
self.spark.column.isNull(),
SF.lit(None),
).otherwise(func(self.spark.column).over(window))
else:
# Here, we use two Windows.
# One for real data.
# The other one for setting nulls after the first null it meets.
#
# There is a behavior difference between pandas and PySpark. In case of cummax,
#
# Input:
# A B
# 0 2.0 1.0
# 1 5.0 NaN
# 2 1.0 0.0
# 3 2.0 4.0
# 4 4.0 9.0
#
# pandas:
# A B
# 0 2.0 1.0
# 1 5.0 NaN
# 2 5.0 NaN
# 3 5.0 NaN
# 4 5.0 NaN
#
# PySpark:
# A B
# 0 2.0 1.0
# 1 5.0 1.0
# 2 5.0 1.0
# 3 5.0 4.0
# 4 5.0 9.0
scol = F.when(
# By going through with max, it sets True after the first time it meets null.
F.max(self.spark.column.isNull()).over(window),
# Manually sets nulls given the column defined above.
SF.lit(None),
).otherwise(func(self.spark.column).over(window))
return self._with_new_scol(scol)
def _cumsum(self, skipna: bool, part_cols: Sequence["ColumnOrName"] = ()) -> "Series":
psser = self
if isinstance(psser.spark.data_type, BooleanType):
psser = psser.spark.transform(lambda scol: scol.cast(LongType()))
elif not isinstance(psser.spark.data_type, NumericType):
raise TypeError(
"Could not convert {} ({}) to numeric".format(
spark_type_to_pandas_dtype(psser.spark.data_type),
psser.spark.data_type.simpleString(),
)
)
return psser._cum(F.sum, skipna, part_cols)
def _cumprod(self, skipna: bool, part_cols: Sequence["ColumnOrName"] = ()) -> "Series":
if isinstance(self.spark.data_type, BooleanType):
scol = self._cum(
lambda scol: F.min(F.coalesce(scol, SF.lit(True))), skipna, part_cols
).spark.column.cast(LongType())
elif isinstance(self.spark.data_type, NumericType):
num_zeros = self._cum(
lambda scol: F.sum(F.when(scol == 0, 1).otherwise(0)), skipna, part_cols
).spark.column
num_negatives = self._cum(
lambda scol: F.sum(F.when(scol < 0, 1).otherwise(0)), skipna, part_cols
).spark.column
sign = F.when(num_negatives % 2 == 0, 1).otherwise(-1)
abs_prod = F.exp(
self._cum(lambda scol: F.sum(F.log(F.abs(scol))), skipna, part_cols).spark.column
)
scol = F.when(num_zeros > 0, 0).otherwise(sign * abs_prod)
if isinstance(self.spark.data_type, IntegralType):
scol = F.round(scol).cast(LongType())
else:
raise TypeError(
"Could not convert {} ({}) to numeric".format(
spark_type_to_pandas_dtype(self.spark.data_type),
self.spark.data_type.simpleString(),
)
)
return self._with_new_scol(scol)
# ----------------------------------------------------------------------
# Accessor Methods
# ----------------------------------------------------------------------
dt = CachedAccessor("dt", DatetimeMethods)
str = CachedAccessor("str", StringMethods)
cat = CachedAccessor("cat", CategoricalAccessor)
plot = CachedAccessor("plot", PandasOnSparkPlotAccessor)
# ----------------------------------------------------------------------
def _apply_series_op(
self, op: Callable[["Series"], Union["Series", Column]], should_resolve: bool = False
) -> "Series":
psser_or_scol = op(self)
if isinstance(psser_or_scol, Series):
psser = psser_or_scol
else:
psser = self._with_new_scol(cast(Column, psser_or_scol))
if should_resolve:
internal = psser._internal.resolved_copy
return first_series(DataFrame(internal))
else:
return psser.copy()
def _reduce_for_stat_function(
self,
sfun: Union[Callable[[Column], Column], Callable[[Column, DataType], Column]],
name: str_type,
axis: Optional[Axis] = None,
numeric_only: bool = True,
**kwargs: Any
) -> Scalar:
"""
Applies sfun to the column and returns a scalar
Parameters
----------
sfun : the stats function to be used for aggregation
name : original pandas API name.
axis : used only for sanity check because series only support index axis.
numeric_only : not used by this implementation, but passed down by stats functions
"""
from inspect import signature
axis = validate_axis(axis)
if axis == 1:
raise ValueError("Series does not support columns axis.")
num_args = len(signature(sfun).parameters)
spark_column = self.spark.column
spark_type = self.spark.data_type
if num_args == 1:
# Only pass in the column if sfun accepts only one arg
scol = cast(Callable[[Column], Column], sfun)(spark_column)
else: # must be 2
assert num_args == 2
# Pass in both the column and its data type if sfun accepts two args
scol = cast(Callable[[Column, DataType], Column], sfun)(spark_column, spark_type)
min_count = kwargs.get("min_count", 0)
if min_count > 0:
scol = F.when(Frame._count_expr(spark_column, spark_type) >= min_count, scol)
result = unpack_scalar(self._internal.spark_frame.select(scol))
return result if result is not None else np.nan
# Override the `groupby` to specify the actual return type annotation.
def groupby(
self,
by: Union[Name, "Series", List[Union[Name, "Series"]]],
axis: Axis = 0,
as_index: bool = True,
dropna: bool = True,
) -> "SeriesGroupBy":
return cast(
"SeriesGroupBy", super().groupby(by=by, axis=axis, as_index=as_index, dropna=dropna)
)
groupby.__doc__ = Frame.groupby.__doc__
def _build_groupby(
self, by: List[Union["Series", Label]], as_index: bool, dropna: bool
) -> "SeriesGroupBy":
from pyspark.pandas.groupby import SeriesGroupBy
return SeriesGroupBy._build(self, by, as_index=as_index, dropna=dropna)
def __getitem__(self, key: Any) -> Any:
try:
if (isinstance(key, slice) and any(type(n) == int for n in [key.start, key.stop])) or (
type(key) == int
and not isinstance(self.index.spark.data_type, (IntegerType, LongType))
):
# Seems like pandas Series always uses int as positional search when slicing
# with ints, searches based on index values when the value is int.
return self.iloc[key]
return self.loc[key]
except SparkPandasIndexingError:
raise KeyError(
"Key length ({}) exceeds index depth ({})".format(
len(key), self._internal.index_level
)
)
def __getattr__(self, item: str_type) -> Any:
if item.startswith("__"):
raise AttributeError(item)
if hasattr(MissingPandasLikeSeries, item):
property_or_func = getattr(MissingPandasLikeSeries, item)
if isinstance(property_or_func, property):
return property_or_func.fget(self) # type: ignore
else:
return partial(property_or_func, self)
raise AttributeError("'Series' object has no attribute '{}'".format(item))
def _to_internal_pandas(self) -> pd.Series:
"""
Return a pandas Series directly from _internal to avoid overhead of copy.
This method is for internal use only.
"""
return self._psdf._internal.to_pandas_frame[self.name]
def __repr__(self) -> str_type:
max_display_count = get_option("display.max_rows")
if max_display_count is None:
return self._to_internal_pandas().to_string(name=self.name, dtype=self.dtype)
pser = self._psdf._get_or_create_repr_pandas_cache(max_display_count)[self.name]
pser_length = len(pser)
pser = pser.iloc[:max_display_count]
if pser_length > max_display_count:
repr_string = pser.to_string(length=True)
rest, prev_footer = repr_string.rsplit("\n", 1)
match = REPR_PATTERN.search(prev_footer)
if match is not None:
length = match.group("length")
dtype_name = str(self.dtype.name)
if self.name is None:
footer = "\ndtype: {dtype}\nShowing only the first {length}".format(
length=length, dtype=pprint_thing(dtype_name)
)
else:
footer = (
"\nName: {name}, dtype: {dtype}"
"\nShowing only the first {length}".format(
length=length, name=self.name, dtype=pprint_thing(dtype_name)
)
)
return rest + footer
return pser.to_string(name=self.name, dtype=self.dtype)
def __dir__(self) -> Iterable[str_type]:
if not isinstance(self.spark.data_type, StructType):
fields = []
else:
fields = [f for f in self.spark.data_type.fieldNames() if " " not in f]
return list(super().__dir__()) + fields
def __iter__(self) -> None:
return MissingPandasLikeSeries.__iter__(self)
if sys.version_info >= (3, 7):
# In order to support the type hints such as Series[...]. See DataFrame.__class_getitem__.
def __class_getitem__(cls, params: Any) -> Type[SeriesType]:
return _create_type_for_series_type(params)
elif (3, 5) <= sys.version_info < (3, 7):
# The implementation is in its metaclass so this flag is needed to distinguish
# pandas-on-Spark Series.
is_series = None
def unpack_scalar(sdf: SparkDataFrame) -> Any:
"""
Takes a dataframe that is supposed to contain a single row with a single scalar value,
and returns this value.
"""
l = cast(pd.DataFrame, sdf.limit(2).toPandas())
assert len(l) == 1, (sdf, l)
row = l.iloc[0]
l2 = list(row)
assert len(l2) == 1, (row, l2)
return l2[0]
@overload
def first_series(df: DataFrame) -> Series:
...
@overload
def first_series(df: pd.DataFrame) -> pd.Series:
...
def first_series(df: Union[DataFrame, pd.DataFrame]) -> Union[Series, pd.Series]:
"""
Takes a DataFrame and returns the first column of the DataFrame as a Series
"""
assert isinstance(df, (DataFrame, pd.DataFrame)), type(df)
if isinstance(df, DataFrame):
return df._psser_for(df._internal.column_labels[0])
else:
return df[df.columns[0]]
def _test() -> None:
import os
import doctest
import sys
from pyspark.sql import SparkSession
import pyspark.pandas.series
os.chdir(os.environ["SPARK_HOME"])
globs = pyspark.pandas.series.__dict__.copy()
globs["ps"] = pyspark.pandas
spark = (
SparkSession.builder.master("local[4]").appName("pyspark.pandas.series tests").getOrCreate()
)
(failure_count, test_count) = doctest.testmod(
pyspark.pandas.series,
globs=globs,
optionflags=doctest.ELLIPSIS | doctest.NORMALIZE_WHITESPACE,
)
spark.stop()
if failure_count:
sys.exit(-1)
if __name__ == "__main__":
_test()