Wilson score calculation for anime titles on shikimori.org

shikimori_ratings.py

#!/usr/bin/env python

import operator
import json

from bs4 import BeautifulSoup
from urllib.request import urlopen
from functools import reduce, lru_cache
from math import sqrt

def sum_prod(lists):
	return sum([reduce(operator.mul, i, 1) for i in zip(*lists)])

# taken from https://habr.com/company/darudar/blog/143188
def _wilson_score(sum_rating, n, votes_range = [0, 1]):
	z = 1.64485
	v_min = min(votes_range)
	v_width = float(max(votes_range) - v_min)
	phat = (sum_rating - n * v_min) / v_width / float(n)
	rating = (phat+z*z/(2*n)-z*sqrt((phat*(1-phat)+z*z/(4*n))/n))/(1+z*z/n)
	return rating * v_width + v_min
  
def wilson_score(votes, votes_range = [1, 10]):
	sum_rating = sum_prod((votes, range(votes_range[0], votes_range[-1] + 1) ))
	n = sum(votes)
	if n == 0:
		return 0
	return _wilson_score(sum_rating, n, votes_range)

def my_score(votes, threshold = 300):
	q = 1/1024., 2/1024., 4/1024., 8/1024., 16/1024., \
	    32/1024., 64/1024., 128/1024., 256/1024., 512/1024.
	norm = 2 / 3.
	n = sum(votes)
	if n < threshold:
		return 0
	return norm * sum_prod((votes, range(1, 11), [i + 1 for i in q])) / n

@lru_cache(maxsize=None)
def get_stats(url):
	res = [0] * 10
	page = BeautifulSoup(urlopen(url).read())
	try:
		stats = json.loads(page.find(attrs = {"id": "rates_scores_stats"}).get("data-stats")
		for stat in stats:
			key, val = stat.values()
			res[key - 1] = val
	except AttributeError:
		pass

	return res


''' Example of usage:
>>> wilson_score(get_stats('https://shikimori.org/animes/9253-steins-gate'))
		 
9.075422968506256'''