Entropy¶

Running entropy.

Parameters¶

fading_factor

Default → 1

Fading factor.
eps

Default → 1e-08

Small value that will be added to the denominator to avoid division by zero.

Attributes¶

entropy (float)

The running entropy.
n (int)

The current number of observations.
counter (collections.Counter)

Count the number of times the values have occurred

Examples¶

import math
import random
import numpy as np
from scipy.stats import entropy
from river import stats

def entropy_list(labels, base=None):
  value,counts = np.unique(labels, return_counts=True)
  return entropy(counts, base=base)

SEED = 42 * 1337
random.seed(SEED)

entro = stats.Entropy(fading_factor=1)

list_animal = []
for animal, num_val in zip(['cat', 'dog', 'bird'],[301, 401, 601]):
    list_animal += [animal for i in range(num_val)]
random.shuffle(list_animal)

for animal in list_animal:
    _ = entro.update(animal)

print(f'{entro.get():.6f}')

1.058093

print(f'{entropy_list(list_animal):.6f}')

1.058093

Methods¶

get

Return the current value of the statistic.

update

Update and return the called instance.

Parameters

x — 'numbers.Number'

Sovdat, B., 2014. Updating Formulas and Algorithms for Computing Entropy and Gini Index from Time-Changing Data Streams. arXiv preprint arXiv:1403.6348. ↩