A Gentle Introduction

An Array is a collection of data just like lists,dictionary,sets and tuples; numpy provides more features and operation on arrays which is widely helpful in feild of data science and analytics. Its also called as a Vector and single element/constants are generally know as scalars in numpy. Array in numpy are in types: 1D array and MultiDimensional array.Generally most people use 2D and 3D arrays in Multidimensional array.

There are many array operations in numpy arrays, they are categorised into 3 types:-

1. Array with Array
2. Array with Scalars
3. Universal Array Functions

There are typically 5 basic arithmetic operations in the first 2 categories mentioned above. They are Addition, Subtraction, Multiplication, Division and Remainder operations.
Note: This article only covers from (Array with Array) operations to (Array with scalar) operations, Universal Array Function will be discussed in upcoming article 'Universal Array Function'.

Prerequisites: 

Just knowledge of basic python programming and concepts of array should be known before learning this concept.
SOFTWARES REQUIRED: A pc with python and numpy installed and any IDE like vscode, pycharm or jupyter notebook is required.
Lets take an example with arr1 and arr2 with values [0,1,2] and [1,2,3] and perform the operations.

Note: Arithmetic operations in arrays can be done on operands with similar shape/size or scalars. A way to overcome this issue is Broadcasting

Addition (operator: '+')

1. With Array and Array:
   Similar to matric addition, this operation follows result[i]=arr1[i]+arr2[i] principle.The other operation follows same principle but with their respective operators ( - , / , % , * )

   import numpy as np
   arr1= np.array([0,1,2])
   arr2= np.array([1,2,3])
   print(arr1+arr2)
   

   Output: [1 3 5]
   return value of arr1+arr2: array([1, 3, 5])

2. With Scalars:
   The shape of scalars are represented as (1,1) and arr1 is represented as (1,3) in 2D vector representation.
   Addition is done in a such a way that scalar is added to every element in array, this is followed as same in all other arithmetic operations in scalar-array operations.

   import numpy as np
   arr1= np.array([0,1,2])
   # scalar '1' will be added to every element in arr1
   print(arr1+1)
   

   Output: [1 2 3]
   return value of arr1+1: array([1, 2, 3])

Subtraction (operator: '-')

1. With Array and Array:

   import numpy as np
   arr1= np.array([0,1,2])
   arr2= np.array([1,2,3])
   print(arr1-arr2)
   

   Output: [-1 -1 -1]
   return value of arr1-arr2: array([-1, -1, -1])

2. With Scalars:
   

   import numpy as np
   arr1= np.array([0,1,2])
   # scalar '1' will be subtracted from every element in arr1
   print(arr1-1)
   

   Output: [-1 0 1]
   return value of arr1-1: array([-1, 0, 1])

Multiplication (operator: '*')

1. With Array and Array:

   import numpy as np
   arr1= np.array([0,1,2])
   arr2= np.array([1,2,3])
   print(arr1*arr2)
   

   Output: [0 2 6]
   return value of arr1*arr2: array([0, 2, 6])

2. With Scalars:
   

   import numpy as np
   arr1= np.array([0,1,2])
   print(arr1*2)
   

   Output: [0 2 4]
   return value of arr1*2: array([0, 2, 4])

Division (operator: '/')

1. With Array and Array:

   import numpy as np
   arr1= np.array([2,2,1])
   arr2= np.array([1,2,2])
   print(arr1/arr2)
   

   Output: [2. 1. 0.5]
   return value of arr1/arr2: array([2.0, 1.0, 0.5])

   • Note: There are 2 exceptional cases in division operation where python throws an error but numpy doesnt, usually in numpy 0/0 returns 'nun' and any number with divident 0 i.e; number/0 returns 'inf' (infinity)*
   • IMP: To be noted that division operator returns resulting values in float64 datatype*
2. With Scalars:
   

   import numpy as np
   arr= np.array([2,3,4])
   print(arr/2)
   

   Output: [1. 1.5 2.]
   return value of arr1/2: array([1.0, 1.5, 2.0])

Remainder (operator: '%')

1. With Array and Array:

   import numpy as np
   arr1= np.array([4,5,9])
   arr2= np.array([2,2,3])
   print(arr1%arr2)
   

   Output: [0 1 0]
   return value of arr1%arr2: array([0, 1, 0])

2. With Scalars:
   

   import numpy as np
   arr1= np.array([100,125,139])
   print(arr1%10)
   

   Output: [0 5 9]
   return value of arr1%10: array([0, 5, 9])

Special operation:

Exponential operation (operator: '**')

Syntax:
base_array**exponent_array

1. With Array and Array:

   import numpy as np
   arr1= np.array([1,2,3])
   arr2= np.array([1,2,3])
   print(arr1**arr2)
   

   Output: [1 4 27]
   return value of arr1** arr2: array([1, 4, 27])

2. With Scalars:
   

   import numpy as np
   arr1= np.array([1,2,3])
   print(arr1**2)
   

   Output: [1 4 9]
   return value of arr1** 2: array([1, 4, 9])

Conclusion:

There are still more operations on arrays and it will be discussed in my upcomming blog articles, stay tuned....
Well thank you very much for your time 👩‍💻
Bye ~(=^‥^)ノ👋
FOLLOW Me: likhith usurupati
LinkedIn: @Likhith