andlima

def distance(points, i, j):
    x1, y1 = points[i]
    x2, y2 = points[j]
    return ((x1 - x2) ** 2 + (y1 - y2) ** 2) ** 0.5

def merge_root(root, i, j):
    good, bad = sorted([root[i], root[j]])
    bad_keys = [k for k, v in root.iteritems() if v == bad]
    for key in bad_keys:
        root[key] = good

andlima

def distance(points, i, j):
    x1, y1 = points[i]
    x2, y2 = points[j]
    return ((x1 - x2) ** 2 + (y1 - y2) ** 2) ** 0.5

def find_root(height_and_parent, i):
    hi, pi = height_and_parent[i]
    if pi is None:
        return i
    return find_root(height_and_parent, pi)

andlima

from bisect import bisect_left, bisect_right


def solve(numbers, a, b):
    answer = bisect_right(numbers, b) - bisect_left(numbers, a)
    return answer if answer > 0 else 0

def main():
    numbers = [1, 4, 9]
    previous = ('012', [''])

andlima

from __future__ import print_function


class Trie(object):
    '''Implementation of a trie.'''

    def __init__(self, collection=None):
        self.ends = False
        self.children = {}
        if collection is not None:

andlima

fn initialize_generator(mt: &mut [uint], index: &mut uint, seed: uint)
{
    mt[0] = seed;
    *index = 0;

    for i in range(1, 624) {
        mt[i] = (
            (
                1812433253 * (mt[i-1] ^ (mt[i-1] >> 30))
            ) + i

andlima

andlima

Table of Contents

• 1. Princípios de Modelagem Preditiva
• 2. Geral
  • 2.1. "Todos os modelos são errados; alguns são úteis"
  • 2.2. Tenha clareza sobre as premissas do modelo
  2.3. Considere o trade-off acurácia vs interpretabilidade