Skip to content

Instantly share code, notes, and snippets.

View ijkilchenko's full-sized avatar
👽

Alex Ilchenko ijkilchenko

👽
View GitHub Profile
# author: @ijkilchenko
# MIT License
def words_over_vocab(vocab, n):
return _words_over_vocab(vocab, '', n)
def _words_over_vocab(vocab, word, n):
if len(word) == n:
yield word
else:
#!/usr/bin/python3
# author: @ijkilchenko
# MIT License
import math
import hashlib
import random
from collections import defaultdict
def base10_to_baseB(num, B):
import random
import math
from hashlib import md5
class MyHashTable(object):
def __init__(self, hash_functions, len_table):
self.hash_functions = hash_functions
self.len_table = len_table
self.table = [-1]*self.len_table
@ijkilchenko
ijkilchenko / heap.py
Last active June 16, 2016 04:19
Heap which supports insertion in O(log(n)) time and finding the max (the root node) in O(1). Also, the tree of the heap is always as short as possible -- what I call a balanced heap.
import random
class Node():
def __init__(self, data=None):
self.data = data
self.parent = None
self.left = None
self.right = None
self.load = 1 # Assumed to be a leaf when initialized.
import random
import numpy as np
#from cvxpy import *
def p(M):
"""Method to pretty print our matrix. """
for i in range(len(M)):
print(' '.join([str(num) for num in M[i]]))
def generate_M(N=3, max_touches=2):
@ijkilchenko
ijkilchenko / exchange_rates.py
Last active April 23, 2023 08:19
Python script which makes a matrix of exchange rates (USD to EUR, etc.), deletes some entries, and then fills-in the gaps (as much as possible)
import random
import copy
def make_exchange_matrix(N=3, seed=2017):
'''In this function, we make an exchange matrix
for testing purposes by randomly picking exchange
rates from the first currency to all the other
currencies. '''
syntax enable
colorscheme monokai
set number
set ruler
set swapfile
set dir=~/tmp
set tabstop=4
from sklearn.datasets import make_classification
from sklearn.cross_validation import train_test_split
from itertools import product
import numpy as np
import pandas as pd
class NeuralNetwork:
g = np.vectorize(lambda z: 1 / (1 + np.exp(-z)))
log = np.vectorize(lambda x: np.log(x))
from math import log10, floor
def r(num):
num_digits = floor(log10(num) + 1)
return num//10 + (num % 10)*(10**(num_digits-1))
def does_a_divide_b(a, b):
if b % a == 0:
return True
else:
import random
from math import sqrt, floor
"""
Guessing game. When playing, the game picks a random value from i to j
and each time a guess g is made, g is added to the previous running cost
of playing this game.
"""
cache = {}
class Game: