3.10. Series Arithmetic

3.10.1. SetUp

>>> import pandas as pd
>>> import numpy as np

3.10.2. Vectorized Operations

  • s + 2, s.add(2), s.__add__(2)

  • s - 2, s.sub(2), s.subtract(2), s.__sub__(2)

  • s * 2, s.mul(2), s.multiply(2), s.__mul__(2)

  • s ** 2, s.pow(2), s.__pow__(2)

  • s ** (1/2), s.pow(1/2), s.__sub__(1/2)

  • s / 2, s.div(2), s.divide(), s.__div__(2)

  • s // 2, s.truediv(2), s.__truediv__(2)

  • s % 2, s.mod(2), s.__mod__(2)

  • divmod(s, 2), s.divmod(2), s.__divmod__(2), (s//2, s%2)

>>> data = pd.Series(
...     data = [1.1, 2.2, np.nan, 4.4],
...     index = ['a', 'b', 'c', 'd'])
>>> data
a    1.1
b    2.2
c    NaN
d    4.4
dtype: float64
>>> data + 2
a    3.1
b    4.2
c    NaN
d    6.4
dtype: float64
>>> data ** 2
a     1.21
b     4.84
c      NaN
d    19.36
dtype: float64
>>> data ** (1/2)
a    1.048809
b    1.483240
c         NaN
d    2.097618
dtype: float64

3.10.3. Broadcasting

  • Uses inner join

  • fill_value: If data in both corresponding Series locations is missing the result will be missing

>>> a = pd.Series([1, 2, 3])
>>> b = pd.Series([4, 5, 6])
>>>
>>> a + b
0    5
1    7
2    9
dtype: int64
>>> a = pd.Series([1, 2, 3, 4])
>>> b = pd.Series([4, 5, 6])
>>>
>>> a + b
0    5.0
1    7.0
2    9.0
3    NaN
dtype: float64
>>> a = pd.Series([1, 2, 3])
>>> b = pd.Series([4, 5, 6, 7])
>>>
>>> a + b
0    5.0
1    7.0
2    9.0
3    NaN
dtype: float64
>>> a = pd.Series([1, 2, None])
>>> b = pd.Series([4, 5, 6])
>>>
>>> a + b
0    5.0
1    7.0
2    NaN
dtype: float64
>>> a = pd.Series([1, 2, None])
>>> b = pd.Series([4, 5, None])
>>>
>>> a + b
0    5.0
1    7.0
2    NaN
dtype: float64
>>> a = pd.Series(data=[1, 2, 3], index=['a', 'b', 'c'])
>>> b = pd.Series(data=[4, 5, 6], index=['a', 'b', 'x'])
>>>
>>> a + b
a    5.0
b    7.0
c    NaN
x    NaN
dtype: float64

fill_value: If data in both corresponding Series locations is missing the result will be missing:

>>> a = pd.Series(data=[1, 2, 3], index=['a', 'b', 'c'])
>>> b = pd.Series(data=[4, 5, 6], index=['a', 'b', 'x'])
>>>
>>> a.add(b, fill_value=0)
a    5.0
b    7.0
c    3.0
x    6.0
dtype: float64

3.10.4. Assignments

Code 3.55. Solution
"""
* Assignment: Series Arithmetic
* Complexity: easy
* Lines of code: 5 lines
* Time: 5 min

English:
    1. Set random seed to zero
    2. Generate `data: ndarray` with 5 random digits [0, 9]
    3. Create `index: list` with index names as sequential letters in english alphabet
    4. Create `s: pd.Series` from `data` and `index`
    5. Multiply `s` by 10
    6. Multiply `s` by `s`
    7. Run doctests - all must succeed

Polish:
    1. Ustaw random ziarno losowości na zero
    2. Wygeneruj `data: np.ndarray` z 5 losowymi cyframi <0, 9>
    3. Stwórz `index: list` z indeksami jak kolejne listery alfabetu angielskiego
    4. Stwórz `s: pd.Series` z `data` oraz `index`
    5. Pomnóż `s` przez 10
    6. Pomnóż `s` przez  wartości `s`
    7. Uruchom doctesty - wszystkie muszą się powieść

Tests:
    >>> import sys; sys.tracebacklimit = 0

    >>> assert result is not Ellipsis, \
    'Assign result to variable: `result`'
    >>> assert type(result) is pd.Series, \
    'Variable `result` has invalid type, should be `pd.Series`'

    >>> result
    a    2500
    b       0
    c     900
    d     900
    e    4900
    dtype: int64
"""

import pandas as pd
import numpy as np
np.random.seed(0)


# type: pd.Series
result = ...