NumPy quantile()

import numpy as np

# create an array
array1 = np.array([0, 1, 2, 3, 4, 5, 6, 7])

# calculate the 0.25th, 0.50th and 0.75th quantile of the array
q25 = np.quantile(array1, 0.25)
q50 = np.quantile(array1, 0.50)
q75 = np.quantile(array1, 0.75)

print(q25, q50, q75)

# Output: 1.75 3.5 5.25

quantile() Syntax

The syntax of the numpy.quantile() method is:

numpy.quantile(array, q, axis = None, out = None, overwrite_input = False, method = 'linear', keepdims = False, interpolation = None)

quantile() Arguments

The numpy.quantile() method takes the following arguments:

• array - input array (can be array_like)
• q - qth quantile to find (can be array_like of float)
• axis (optional) - axis or axes along which the quantiles are computed (int or tuple of int)
• out (optional) - output array in which to place the result (ndarray)
• keepdims (optional) - specifies whether to preserve the shape of the original array (bool)
• override_input (optional) - bool value that determines if intermediate calculations can modify an array
• method (optional) - the interpolation method to use
• interpolation (optional) - the deprecated name for the method keyword argument

quantile() Return Value

The numpy.quantile() method returns the q-th quantile(s) of the input array along the specified axis.

Quantile

The quantile is a statistical measure that represents the value below which a specific percentage of data falls. It helps analyze the distribution of a dataset.

In NumPy, the quantile() function computes the q-th quantile of data along the specified axis.

The q-th quantile represents the value below which q percent of the data falls. For example, the 0.50th quantile (also known as the median) divides the data into two equal halves.

Note: numpy.quantile() and numpy.percentile() do the same thing. If you want to specify q from 0 to 100, use percentile() and if you want to specify q from 0.0 to 1.0, use quantile().

Example 1: Find the Quantile of an ndArray

import numpy as np

# create an array
array1 = np.array([[[0, 1],
                    [2, 3]],

                   [[4, 5],
                    [6, 7]]])

# find the 50th quantile of entire array
quantile1 = np.quantile(array1, q = 0.50)

# find the 50th quantile across axis 0
quantile2 = np.quantile(array1, q = 0.50, axis = 0)

# find the 50th quantile across axis 0 and 1
quantile3 = np.quantile(array1, q = 0.50, axis = (0, 1))

print('\n50th quantile of the entire array:', quantile1)
print('\n50th quantile across axis 0:\n', quantile2)
print('\n50th quantile across axis 0 and 1:', quantile3)

Output

50th quantile of the entire array: 3.5

50th quantile across axis 0:
[[2. 3.]
 [4. 5.]]

50th quantile across axis 0 and 1: [3. 4.]

Example 2: Using Optional out Argument

The out parameter allows us to specify an output array where the result will be stored.

import numpy as np

arr = np.array([[1, 2, 3],
                [4, 5, 6]])

# create an output array
output = np.zeros(3)

# compute 25th quantile and store the result in the output array
np.quantile(arr, 0.25, out = output, axis = 0)

print('25th quantile:', output)

Output

25th quantile: [1.75 2.75 3.75]

Example 3: Using Optional keepdims Argument

If keepdims is set to True, the resultant array's dimensions are the same as the original array.

import numpy as np

arr = np.array([[1, 2, 3],
                [4, 5, 6]])

# keepdims defaults to False
result1 = np.quantile(arr, 0.50 , axis = 0)

# pass keepdims as True
result2 = np.quantile(arr, 0.50, axis = 0, keepdims = True)

print('Dimensions in original array:', arr.ndim)
print('Without keepdims:', result1, 'with dimensions', result1.ndim)
print('With keepdims:', result2, 'with dimensions', result2.ndim)

Output

Dimensions in original array: 2
Without keepdims: [2.5 3.5 4.5] with dimensions 1
With keepdims: [[2.5 3.5 4.5]] with dimensions 2