💡Why Numpy ?
1️⃣ Numerical Operations are faster than normal python lists
2️⃣ Also known as Array Oriented Computing
3️⃣ Unlike lists, Numpy Array elements should be of same data type
4️⃣ package for creating Multi-Dimension Arrays
some info on Arrays
18 is a Scalar (there is no dimension, so no Shape)
1️⃣ D Array = Vector (collection of scalars) [1, 2, 3, 4, 5]
It has 1 axes i.e., axis = 0
Shape is (N, ) - N is no. of scalars
here, Shape is (5, )
2️⃣ D Array = Matrix (collection of Vectors) [[1, 2, 3, 4, 5], [6, 7, 8, 9, 10]]
It has 2 axes i.e., axis=0, axis=1
Shape is (N, M) - N is no. of vectors, M is no of scalars in each vector
here, Shape is (2, 5)
3️⃣ D Array (collection of Matrix's)
[[[1, 2, 3, 4, 5],
[6, 7, 8, 9, 10]],
[[11, 12, 13, 14, 15],
[16, 17, 18, 19, 20]]]
anything more than 2D array can be called Tensor
Shape is (N, M, P) - N number of Matrices each of shape (M, P)
here, Shape is (2, 3, 5)
💡 How to Create an Array ?
- 3 ways
import numpy as np
➖ np.array(your_list)
your list or nested list to create 1D/higher dimension
arrays
➖ np.arange(5) - creates 1D Array
read it as a
range (range() in python)
➖ np.random.rand(5)
💡 Understand our Array
a = np.array([1,2,3,4])
→ a.dtype # data type is dtype('int64') → a.dim # dimension is 1 → a.shape # shape is (4, ) → a.size # total number of elements
If you like the content, you could follow on twitter @nameismahipal