Distinctive Analytics

Features

• It also has functions for matrices, linear algebra, and Fourier Transform.
• Python has lists which do the work of an array, but numpy will provide 50x more speed than a Python list.
• Numpy is written in c,c++ and Python.
• Where fast computations are required c,c++ is used.
• It works efficiently with the latest CPU architecture.

Numpy Installation

• pip install numpy
• Using this command will ensure the installation of the latest version of numpy that is compatible with your system.

pip install numpy

Importing Numpy into your code

import numpy

import numpy as np

Numpy datatypes

import numpy as np
pi = np.float32(3.16)
print(type(pi))
print(pi)

<class 'numpy.float32'>
3.16

Create Ndarray with Numpy

num_arr = np.array(42)

print(type(num_arr))

print(num_arr)

Create 1 Dimensional Array

• This array has a single dimension.
• It is a collection of a single unit.

num_arr = np.array([1, 2, 3, 4, 5])
print(num_arr)
print(type(num_arr))
print(num_arr.shape)
print(num_arr.ndim)

[1 2 3 4 5]
<class 'numpy.ndarray'>
(5,)
1

Create 2 Dimensional Array

• An array that hosts multiple 1 Dimensional arrays is a 2 Dimensional array
• 2 Dimensional arrays store matrices, tensors and vectors.

num_arr1=np.array([1, 2, 3, 4, 5])

num_arr2=np.array([6, 7, 8, 9, 10])

TwoD = np.array( [ num_arr1 , num_arr2 ] ) 

print(TwoD)

print(TwoD.shape)

print(TwoD.ndim)

[[ 1  2  3  4  5]
 [ 6  7  8  9 10]]
(2, 5)
2

Create 3 Dimensional Array

• An array that hosts multiple-dimensional arrays is a 3 3-dimensional array.
• 3D arrays are used to represent 3rd-order Tensor.

num_arr1=np.array([1, 2, 3])
num_arr2=np.array([4, 5, 6])
num_arr3=np.array([7, 8, 9])
T3D=np.array([
              [num_arr1,num_arr2],
 [num_arr2,num_arr3],
              [num_arr1,num_arr3]
              ]
             ) 
print(T3D)
print(T3D.shape)
print(T3D.ndim)

[[[1 2 3]
  [4 5 6]]

 [[4 5 6]
  [7 8 9]]

 [[1 2 3]
  [7 8 9]]]
(3, 2, 3)
3

Higher Dimensions

• We can create multiple dimensions in a Numpy Array.
• We either define multidimensional arrays or we can provide them already formatted.

ndim

Furthermore, we will use ndim attribute to determine the number of dimensions.

np.array([[[[1,2,3], [4,5,6], [7,8,9]], [[10,11,12], [13,14,15], [16,17,18]]], [[[19,20,21], [22,23,24], [25,26,27]], [[28,29,30], [31,32,33], [34,35,36]]]])
print(a.ndim)
a

4
array([[[[ 1,  2,  3],
         [ 4,  5,  6],
         [ 7,  8,  9]],

        [[10, 11, 12],
         [13, 14, 15],
         [16, 17, 18]]],


       [[[19, 20, 21],
         [22, 23, 24],
         [25, 26, 27]],

        [[28, 29, 30],
         [31, 32, 33],
         [34, 35, 36]]]])

Shape

Attribute shape determines the length of each dimension inside the whole array since we know the 4 dimensions are created in the array.

import numpy as np
e = np.array([[[[4, 2, 3], [4, 5, 6]], [[1, 2, 3], [4, 5, 6]]]])
e.shape

(2, 2, 3, 3)

Concatenate

By using concatenate, we can combine two arrays along a given axis. In the following examples, we will be showcasing two examples where axis value is set to 0 by default, and we have to set any other value manually.

import numpy as np
arr1 = np.array([1.1, 2.1, 3.1])
arr2 = np.array([1.2, 2.2, 3.2])
arr = np.concatenate((arr1, arr2))
arr

array([1.1, 2.1, 3.1, 1.2, 2.2, 3.2])

import numpy as np
arr1 = np.array([[1.1, 2.1, 3.1],[1.2, 2.2, 3.2],[1.3, 2.3, 3.3]])
arr2 = np.array([[1,2,3]])
arr = np.concatenate((arr1, arr2.T),axis=1)
arr

array([[1.1, 2.1, 3.1, 1. ],
       [1.2, 2.2, 3.2, 2. ],
       [1.3, 2.3, 3.3, 3. ]])

Stack

The Numpy stack allows us to join multiple arrays. And we can also use an axis with this method similar to concatenate.

import numpy as np
arr1 = np.array([1, 2, 3,4])
arr2 = np.array([4, 5, 6,7])
arr3 = np.array([7, 8, 9,10])
arr4 = np.array([10, 11, 12,13])
arr4 = np.array([11, 12, 13,14])
arr = np.stack((arr1, arr2 , arr3 , arr4 ),axis=0)
print(arr)

[[ 1  2  3  4]
 [ 4  5  6  7]
 [ 7  8  9 10]
 [11 12 13 14]]

import numpy as np
arr1 = np.array([1, 2, 3,4])
arr2 = np.array([4, 5, 6,7])
arr3 = np.array([7, 8, 9,10])
arr4 = np.array([10, 11, 12,13])
arr4 = np.array([11, 12, 13,14])
arr = np.stack((arr1, arr2 , arr3 , arr4 ),axis=1)
print(arr)

[[ 1  4  7 11]
 [ 2  5  8 12]
 [ 3  6  9 13]
 [ 4  7 10 14]]

import numpy as np
arr1 = np.array([1, 2, 3,4])
arr2 = np.array([4, 5, 6,7])
arr3 = np.array([7, 8, 9,10])
arr4 = np.array([10, 11, 12,13])
arr4 = np.array([11, 12, 13,14])
arr = np.stack((arr1, arr2 , arr3 , arr4 ),axis=-1)
print(arr)

[[ 1  4  7 11]
 [ 2  5  8 12]
 [ 3  6  9 13]
 [ 4  7 10 14]]

Also, take note of axis 1 and axis-1 output

import numpy as np
arr1 = np.array([[1, 2, 3],[1, 2, 3],[1, 2, 3]])
arr2 = np.array([[4, 5, 6],[4, 5, 6],[4, 5, 6]])
arr = np.dstack(( arr1 , arr2 ))
print(arr)

[[[1 4]
  [2 5]
  [3 6]]

 [[1 4]
  [2 5]
  [3 6]]

 [[1 4]
  [2 5]
  [3 6]]]

import numpy as np
arr1 = np.array([[1, 2, 3],[1, 2, 3],[1, 2, 3]])
arr2 = np.array([[4, 5, 6],[4, 5, 6],[4, 5, 6]])
arr = np.hstack(( arr1 , arr2 ))
print(arr)

[[1 2 3 4 5 6]
 [1 2 3 4 5 6]
 [1 2 3 4 5 6]]
 
 
 
 
 
 
 

import numpy as np
arr1 = np.array([[1, 2, 3],[1, 2, 3],[1, 2, 3]])
arr2 = np.array([[4, 5, 6],[4, 5, 6],[4, 5, 6]])
arr = np.vstack(( arr1 , arr2 ))
print(arr)

[[1 2 3]
 [1 2 3]
 [1 2 3]
 [4 5 6]
 [4 5 6]
 [4 5 6]]
 
 
 
 

import numpy as np
arr1 = np.array([[1, 2, 3],[1, 2, 3],[1, 2, 3]])
arr2 = np.array([[4, 5, 6],[4, 5, 6],[4, 5, 6]])
arr = np.vstack(( arr1 , arr2 ))
print(arr)

[[1 2 3]
 [1 2 3]
 [1 2 3]
 [4 5 6]
 [4 5 6]
 [4 5 6]]