4.3 quantile and digitize

This file describes the concept of quantile and how to cclculate it in numpy and various functions around it.

import numpy as np

print(np.__version__)

1.26.3

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

np.quantile(x, 0.5)

3.0

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

np.quantile(x, 0.5)

3.0

so quantile actually means that what will be that value that if we distribute the values of the array 50% on one side and 50% on other side. 50% because we have used 0.5

x = np.array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])

print(np.quantile(x, 0.5))

4.5

print(np.quantile(x, 0.2))

1.8

print(np.quantile(x, [0.2, 0.5]))

[1.8 4.5]

print(np.quantile(np.arange(10, 20), [0.2, 0.5]))

[11.8 14.5]

what if the array is 2D

print(np.quantile(np.arange(20).reshape(-1, 10), [0.2, 0.5]))

[3.8 9.5]

np.quantile([1.8, 11.8], 0.2)
# 3.8
np.quantile([4.5, 14.5], 0.5)
# 9.5