{{announcement.body}}
{{announcement.title}}

Python - 5 Sets of Useful Numpy Unary Functions - Data Analytics

DZone 's Guide to

Python - 5 Sets of Useful Numpy Unary Functions - Data Analytics

In this post, you will learn about some of the 5 most popular or useful set of unary universal functions (ufuncs) provided by Python Numpy library.

· Big Data Zone ·
Free Resource

In this post, you will learn about some of the 5 most popular or useful set of unary universal functions (ufuncs) provided by Python Numpy library. As data scientists, it will be useful to learn these unary functions by heart as it will help in performing arithmetic operations on sequential-like objects. These functions can also be termed as vectorized wrapper functions which are used to perform element-wise operations.

The following represents different set of popular functions:

  • Basic arithmetic operations
  • Summary statistics
  • Sorting
  • Minimum/maximum
  • Array equality

Basic Arithmetic Operations

The following are some of the unary functions whichc an be used to perform arithmetic operations:

  • add, subtract, multiply, divide, exp
  • add.reduce
  • sum

Here is the sample code demonstrating the usage of the above functions:

Java
 




x
13


 
1
import numpy as np
2
#
3
# Create an array of 5 numbers between 5 and 10
4
#
5
arr = np.linspace(5, 10, 5)
6
#
7
# Print array
8
#
9
print(arr)
10
#
11
# Perform arithmetic operations
12
#
13
np.add(arr, 1), np.subtract(arr, 1), np.multiply(arr, 2), np.divide(arr, 2)


Here is how the output would look like:


Fig 1. Numpy Unary Arithmetic Functions


In case, you want to add all the numbers in a row or column and get the output as matrix, functions such as add.reduce or sum is used with axis. In the code sample given below, a 2 x 5 matrix is reduced by rows (axis=1) and columns (axis=0)


Fig 2. Numpy Unary functions to sum rows and columns


Summary Statistics

The following are some of the methods which can be used for calculating summary statistics:

  • mean, median: Finds the mean and median of the data sample respectively.
  • std: Finds the standard deviation of the data
  • var: Finds the variance of the data

Here is the code demonstrating the usage of above functions with SKlearn IRIS dataset.

Java
 




xxxxxxxxxx
1


 
1
import numpy as np
2
from sklearn import datasets
3
 
           
4
iris = datasets.load_iris()
5
 
           
6
np.mean(iris.data[:,0]), np.std(iris.data[:,0]), np.var(iris.data[:,0]), np.median(iris.data[:,0])


Here is how the output will look like:


Fig 2. Numpy unary functions to fund summary statistics


Sorting

The following represents the Numpy unary functions which can be used for sorting the array:

  • sort: Sort the data
  • argsort: Finds the indices of the sorted data

Here is the code demonstrating the usage of above functions with SKlearn IRIS dataset:

Java
 




xxxxxxxxxx
1
10


 
1
import numpy as np
2
from sklearn import datasets
3
 
           
4
iris = datasets.load_iris()
5
 
           
6
iris.data[0:10, 0]
7
 
           
8
np.sort(iris.data[0:10, 0])
9
 
           
10
np.argsort(iris.data[0:10, 0])


Here is how the output will look like:


Fig 3. Numpy unary functions for sorting


Finding Maximum/Minimum

The following are some of the methods which can be used for finding maximum and minimum value from a data array:

  • min: Finds the minimum of the array
  • max: Finds the maximum of the array
  • argmin: Finds the index of the minimum of the array
  • argmax: Finds the index of the maximum of the array

Here is the code demonstrating the usage of above functions with SKlearn IRIS dataset:

Java
 




xxxxxxxxxx
1
10


 
1
import numpy as np
2
from sklearn import datasets
3
 
           
4
iris = datasets.load_iris()
5
 
           
6
iris.data[0:10, 0]
7
 
           
8
np.min(iris.data[0:10, 0]), np.argmin(iris.data[0:10, 0])
9
 
           
10
np.max(iris.data[0:10, 0]), np.argmax(iris.data[0:10, 0])


Here is how the output will look like:

Fig 4. Numpy unary functions for finding minimum and maximum
Topics:
big data, data analytics, numpy, numpy unary functions, python, tutorial

Published at DZone with permission of Ajitesh Kumar , DZone MVB. See the original article here.

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}