Matrices With Numpy

 

Numpy is really good at working with matrix mathematics, and matrix math is everywhere in data science.

It will save you a lot of time if you get acquainted with the numpy library.

Okay, without further ado, let’s get started by importing numpy and creating a sample matrix to work with.

import numpy as np

example_matrix = np.array([
  [1,2,3,4],
  [5,6,7,8],
  [9,10,11,12],
  [13,14,15,16]
])

One of the first things that we might want to do is get more info on our matrix.

Let’s first take a look at its shape. This will tell us the number of rows by the number of columns.

example_matrix.shape
# (4, 4)

Next, let’s take a look at how many elements are in our matrix.

example_matrix.size
# 16

Then, what if we wanted to know how many dimensions our data was? We got a function for that too.

example_matrix.ndim
# 2

Next, let’s get some statistical information on our matrix. Common statistical properties of a matrix are: min, max, mean, variance, and standard deviation.

np.min(example_matrix)
# 1

np.max(example_matrix)
# 16

np.mean(example_matrix)
# 8.5

np.var(example_matrix)
# 21.25

np.std(example_matrix)
# 4.6097722286464435

Sweet! That was a quick introduction to using matrices with numpy.

I hope you learned something new 🙂

Checkout the complete code example below.

import numpy as np

example_matrix = np.array([
  [1,2,3,4],
  [5,6,7,8],
  [9,10,11,12],
  [13,14,15,16]
])

example_matrix.shape
# (4, 4)

example_matrix.size
# 16

example_matrix.ndim
# 2

np.min(example_matrix)
# 1

np.max(example_matrix)
# 16

np.mean(example_matrix)
# 8.5

np.var(example_matrix)
# 21.25

np.std(example_matrix)
# 4.6097722286464435