fbparis

"""
Librement inspiré de https://twitter.com/IllumiReptilien/status/1325009440026877953
anus.png peut être récupéré ici : https://github.com/fbparis/floot/blob/main/anus.png
"""

# uncomment in colab
# !pip install mtcnn

import cv2
import mtcnn

fbparis

class RandomUCI():
    """Random Upper Confidence Interval
    """
    def __init__(self, bandit):
        self.bandit = bandit
        self.arm_count = bandit.arm_count
        self.mean = np.ones(self.arm_count) / 2
        self.variance = np.ones(self.arm_count) / 4
        self.N = np.ones(self.arm_count)

fbparis

#!/usr/local/bin/python3
# -*- coding: utf-8 -*-
"""
"""

__title__ = __file__.rsplit('.', 1)[0].title()
__author__ = '[email protected]'
__license__ = 'IV'

version_info = (0, 0, 1)

fbparis

# -*- coding: utf-8 -*-
"""
New kind of text extractor:
	Phase 1:
	- remove scripts css svg comments canvas etc
	- add space after each html tag
	- strip tags
	- normalize spaces/newlines
	- keep the result
	Phase 2:

fbparis

"""
Use the formula found on https://nookipedia.com/wiki/Compatibility (slightly modified) to compute Animal Crossing New Horizons village compositions optimizing villagers happiness.

Compatibility score between 2 villagers is a grade from 0 (worst) to 4 (best) and village score is the sum of every combinations of 2 villagers score in the village.

You can pass a required list of villagers (using their english ids or french names) to find the ideal villagers to recrute next.

See --help for options. -m3 -r2 are usualy good parameters.

Note that the program never stops so you will have to hit CTRL-C when you're done!

fbparis

from itertools import product, combinations
from random import sample

# 0 1 2 3
# 4 5 6 7
# 8 9 A B
# C D E F

cards = list(product('ABCD', '0123', repeat=1))

fbparis

"""
kik (key to index to key) is a new implentation of indexed trie (https://gist.github.com/fbparis/e114958a4a9be84146a1a579cf677e8d) reducing memory footprint by a factor 7 or more!

The main differences with indexed tries are:
	* Trie nodes are no longer stored in Node objects but in a python list. An internal mechanism use the list to simulate a tree behaviour, without any recursive function.
	* Nodes and values stored in the kik can be compressed with several compression levels:
		* None: no compression at all
		* 0: data are stored as pickle.dumps (memory use divided by 3-4)
		* -1, 1 to 9: pickle.dumps are compressed with zlib
	* Some internal methods can be cached with a custom implentation of a LRU cache.

fbparis

"""
A Python3 indexed trie class.
An indexed trie's key can be any subscriptable object. 
Keys of the indexed trie are stored using a "radix trie", a space-optimized data-structure which has many advantages (see https://en.wikipedia.org/wiki/Radix_tree).
Also, each key in the indexed trie is associated to a unique index which is build dynamically.

Indexed trie is used like a python dictionary (and even a collections.defaultdict if you want to) but its values can also be accessed or updated (but not created) like a list!

Example:
	>>> t = indextrie()

fbparis

import os
from math import factorial
from random import seed, randrange

class PerfectCipher(object):
	"""
	Each byte (0-255) of a message is replaced using a translation table which is a permutation of [0..255].
	After each replacement a new translation table is used using a Linear Congruential Generator to act as a PRNG:
		Next N = (a * N + c) % m
	Where m is 256! (total count of [0..255] permutations) and a and c carefully chosen so that the period of the PRNG is equal to m.

fbparis

import sys
import socket
import zlib
from loremvpn import LoremVPN

if __name__ == '__main__':
	VPN = LoremVPN("demo", "demo")
	HOST, PORT = "localhost", 9999
	data = " ".join(sys.argv[1:]).encode()