import math

import random

import numpy as np

# 표준편차 계산

def standard_deviation(lst):

m = sum(lst) / len(lst)

variance = sum([(value - m) ** 2 for value in lst])

return math.sqrt(variance / len(lst))

rands = [random.random() for _ in range(0, 1000000)]

numpy_rands = np.array(rands)

# %timeit -n 100 np.std(numpy_rands)

# 3.28 ms ± 214 μs per loop (mean ± std. dev. of 7 runs, 100 loops each)

# %timeit -n 100 standard_deviation(rands)

# 131 ms ± 1.03 ms per loop (mean ± std. dev. of 7 runs, 100 loops each)