Faster Hamming Calculator

Motivation

For calculating Hamming distance, an apparent strategy could be:

from numpy import random, sum

variable_number = 20
sample_1 = random.randint(0, 2, size=(int(variable_number),))
sample_2 = random.randint(0, 2, size=(int(variable_number),))
distance = sum(sample_1 != sample_2)

However, in large-scale operations, such as 1,000,000 times, its runtime is difficult to be accepted by users. Hence, this work is specifically designed for large-scale Hamming distance calculation. The results shown that the runtime comparison is our proposed << numpy.sum (at least one-tenth). Detailed design is shown in here.

Comparison of Hamming distance between one sample and all other samples

This function is often used in the task of investigating whether an individual belongs to a population.

You can complete this task by:

from numpy import array
from calculator import hamming_group
sample = array([0, 1, 0])
sample_group = array([[1, 1, 0], [0, 1, 1], [0, 1, 0]])
hamming_group(observed_sample=sample, sample_group=sample_group, threshold=None)
# array([1, 1, 0])

The comparison with the 'numpy.sum' method is as follows:

Here, "x" implies that observation and statistics cannot be carried out in the unit of seconds.

Comparison of Hamming distance between a group of samples

This function is often used to investigate the differences between individuals within a population.

You can do this task through:

from numpy import array
from calculator import hamming_matrix
samples = array([[1, 1, 0], [0, 1, 1], [0, 1, 0]])
hamming_matrix(samples=samples)
# array([[0, 2, 1],
#        [2, 0, 1],
#        [1, 1, 0]])

The comparison with the 'numpy.sum' method is as follows:

Comparison of Hamming distance between two groups of samples.

This function is usually used to investigate the individual differences between two populations.

You can finish this task using:

from numpy import array
from calculator import hamming_matrix
samples_1 = array([[1, 1, 0], [0, 1, 1], [0, 1, 0]])
samples_2 = array([[0, 0, 1], [1, 0, 1]])
hamming_matrix(samples=samples_1, other_samples=samples_2)
# array([[3, 2],
#        [1, 2],
#        [2, 3]])

The comparison with the 'numpy.sum' method of inter population operation is similar as that of intra population operation.