Overview
- The Series class of the python library pandas, implements a one-dimensional, mutable container for storing heterogeneous elements using the numpy.ndarray as the underlying storage.
- In addition to providing a versatile one dimensional container, the pandas.Series provides numerous computational methods for analyzing the data.
- The subtraction operation is a binary operation. A binary operation consumes two values to produce a new value. In case of subtraction between two pandas.Series instances, one element of the Series is subtracted from the another producing a new Series.
- To subtract two pandas.Series instances, the function Series.sub(**)** is used. The subtraction operator “–“ can as well be used for the same purpose. In the similar way to subtract a DataFrame instance from another, the DataFrame.sub() function can be used.
- While using the sub() function explicitly, it is possible to replace any occurrence of None, using the parameter fill_value.
Example
# Example Python program that subtracts one series from another
import pandas as pds
import numpy as np
# Generate ten random integers between 1 and 20
np.random.seed(1);
randomNumbers = np.random.randint(1, 20, 10);
# Create a pandas series
series1 = pds.Series(randomNumbers);
# Generate ten random integers between 1 to 5
np.random.seed(1);
randomNumbers = np.random.randint(1, 5, 10);
# Create a pandas series
series2 = pds.Series(randomNumbers);
# Subtract a pandas Series instance from another
series3 = series1.sub(series2);
print("Series1:");
print(series1);
print("Series2:");
print(series2);
print("Series3:");
print(series3);Output
Series1:
0 6
1 12
2 13
3 9
4 10
5 12
6 6
7 16
8 1
9 17
dtype: int64
Series2:
0 2
1 4
2 1
3 1
4 4
5 2
6 4
7 2
8 4
9 1
dtype: int64
Series3:
0 4
1 8
2 12
3 8
4 6
5 10
6 2
7 14
8 -3
9 16
dtype: int64Example - Subtracting a pandas Series from another while replacing None values with zero
# Example Python program that subtracts one series from another
# where one or more elements are None.
import pandas as pds
import numpy as np
# Create pandas.Series instances
lookAndSaySeries = pds.Series([1, 11, 21, 1211, 111221, 312211]);
series2Subtract = pds.Series([1, 0, None, 2, None, 4]);
# While subtracting replace any None with zero
newSeries = lookAndSaySeries.sub(series2Subtract, fill_value=0);
print("Look and Say Series:");
print(lookAndSaySeries);
print("Series to Subtract:");
print(series2Subtract);
print("New series after subtraction:");
print(newSeries);Output
Look and Say Series:
0 1
1 11
2 21
3 1211
4 111221
5 312211
dtype: int64
Series to Subtract:
0 1.0
1 0.0
2 NaN
3 2.0
4 NaN
5 4.0
dtype: float64
New series after subtraction:
0 0.0
1 11.0
2 21.0
3 1209.0
4 111221.0
5 312207.0
dtype: float64Example
Any python sequence of same length like a Python tuple, a Python list can be subtracted from a pandas.Series, as given in the Python example program below.
# Example Python program that subtracts the elements
# of a tuple from a pandas.Series
import pandas as pds
import numpy as np
numbers = [1, 1, 4, 8, 22, 51];
# Create a pandas.Series from a Python List
series1 = pds.Series(numbers);
# Create a python tuple
toSubtract = (1, 2, 3, 4, 9, 20);
# Subtract a tuple from a series
resultantSeries = series1.sub(toSubtract);
print("Contents of the pandas.Series instance:");
print(series1);
print("Contents of the Python tuple:");
print(series1);
print("Resultant pandas.Series after subtracting the tuple:");
print(resultantSeries);Output
Contents of the pandas.Series instance:
0 1
1 1
2 4
3 8
4 22
5 51
dtype: int64
Contents of the Python tuple:
0 1
1 1
2 4
3 8
4 22
5 51
dtype: int64
Resultant pandas.Series after subtracting the tuple:
0 0
1 -1
2 1
3 4
4 13
5 31
dtype: int64