ybenjo

# -*- coding: utf-8 -*-
require 'natto'
require 'msgpack'
require 'mongo'

texts = ['shirobako★が3巻が見たい', '大久保瑠美さんに会いたい']

trans_freq = Hash.new{|h, k| h[k] = Hash.new{0.0}}

texts.each do |txt|

ybenjo

agqr.rb

これは何

AGQR の放送を保存するスクリプト．

fork 元との違いは

yagays / agqr.rb には

• 31日まである月に翌日の指定が失敗する

というバグが存在する．

ybenjo

# -*- coding: utf-8 -*-
require 'open-uri'
require 'nokogiri'
require 'logger'
require 'time'
require 'set'
require 'pp'

UA = ''

ybenjo

0  a	0.545564
0	b	0.235328
0	c	0.0589112
0	d	0.160196
1	a	0.282141
1	b	0.400634
1	c	0.227655
1	d	0.0895693
2	a	0.0782046
2	b	0.155836

ybenjo

a  x:0.5488135039273248	y:0.7151893663724195
b	x:-0.6027633760716439	y:0.5448831829968969
c	x:-0.4236547993389047	y:-0.6458941130666561
d	x:0.4375872112626925	y:-0.8917730007820798
a	x:0.9636627605010293	y:0.3834415188257777
b	x:-0.7917250380826646	y:0.5288949197529045
c	x:-0.5680445610939323	y:-0.925596638292661
d	x:0.07103605819788694	y:-0.08712929970154071
a	x:0.02021839744032572	y:0.832619845547938
b	x:-0.7781567509498505	y:0.8700121482468192

ybenjo

#include <iostream>
#include <fstream>
#include <sstream>
#include <unordered_map>
#include <vector>
#include <string>
#include <algorithm>
#include <boost/random.hpp>
#include <iomanip>

ybenjo

#include <iostream>
#include <fstream>
#include <sstream>
#include <unordered_map>
#include <vector>
#include <string>
#include <algorithm>

typedef std::pair<int, int> key;

ybenjo

# GMM (generalized Mixture Model) of EM
# usage
# gmm = GMM.new([1, 2, 3, 5, 2, 1, 10, 20, 30, 20], {k: 2})

LIM = 10 ** -5
PROB_LIM = 10 ** -100

def dist(x, mu, sigma)
  prob = 1.0 / (2 * Math::PI * sigma) ** 0.5 * Math::exp(-(x - mu) ** 2 / (2 * sigma))
  # avoid p(x|mu, sigma) = 0.0

ybenjo

# Xiaohui Yan, A biterm topic model for short texts(WWW 2013)
require 'set'

class Biterm
  def initialize(alpha, beta, k)
    @alpha = alpha
    @beta = beta
    @k = k

    @doc_w = { }

ybenjo

# usage: ruby ikoma.rb | pbcopy
require 'nokogiri'
require 'open-uri'
doc = Nokogiri::HTML(open("http://d.hatena.ne.jp/mamoruk/archive?word=&of=#{rand(1550)}", :proxy => nil).read)
entries = (doc/'li.archive.archive-section'/'a'); elem = entries[rand(entries.size)]
puts "#{elem.inner_text} #{elem.attribute('href').value}"