data15b.rb

require 'statsample'


def f2(n)

	n = (n * 3 + 1) / 2 while (n.odd?)
	n /= 2 while (n.even?)
	return n

end

def adv(x, l2)

	x['n'] = x['nb'].to_i(2)
	n1 = n = x['n']
	l = [n]
	while (n >= n1 && n != 1)
		
		n = f2(n)
		l << n
	end
	

	x['ls'] = l.size
	x['ns'] = x['n'].to_s(2).length
	x['h2'] = (x['ls'].to_f / x['ns'] * 50).to_i
	x['d'] = d(x['nb'])
	
	l2 << x
	
	return x
	
end


def stat(l)
	l = [0] if (l.empty?)
        t = t2 = 0
        l.each \
        {
		|x|
		t += x
		t2 += x ** 2
        }
        c = l.size
        a = t.to_f / c
        z = t2.to_f / c - a ** 2
        sd = Math.sqrt(z < 0 ? 0 : z)

	return a, sd, l.max.to_f, l.min.to_f
end

def init(w, f)
	s = (0...w).map { rand(2).to_s }.join
	s[0, 1] = '1' if (f)
	return s
end

def dist(w, f)
	
	l = []
	l2 = []
	f0, f1, f2 = f
	
	20.times { l << adv({'nb' => init(w, f0)}, l2) }
	c = 2e3.to_i

	seen = {}

	i = n = 0
	while (n < c)

		i += 1
		l.sort_by! { |x| -x['ls'] }

		t = f1 ? [l.size, 400].min : l.size / 10
		case i % 3
			when 0
				x = l[rand(t)]
				s = x['nb']
				r = f0 ? (rand(s.length - 1) + 1) : rand(s.length)
				s[r, 1] = (s[r, 1].to_i  ^ 1).to_s
				
			when 1
				x = l[rand(t)]
				y = l[rand(t)]
				sx = x['nb']
				sy = y['nb']
				s = ''
				w.times \
				{
					|j|
					s[j, 1] = (rand(2) == 0) ? sx[j, 1] : sy[j, 1]
				}
				
			when 2
				x = l[rand(t)]
				y = l[rand(t)]
				sx = x['nb']
				sy = y['nb']
				s = ''
				r = rand(w + 1)
				w.times \
				{
					|j|
					s[j, 1] = (j < r) ? sx[j, 1] : sy[j, 1]
				}
		end
		
		next if (s.to_i == 0)
		
		if (f2) then
			next if (seen.member?(s))
			seen[s] = nil
		end
		
		l.pop if (l.size >= 500)
		l << adv({'nb' => s}, l2)
		n += 1
	end

#	$stderr.puts("#{i - n} dups")
	return l2
end

def hist(l)

	h = {}
	l.each \
	{
		|x|
		h[x] = h.fetch(x, 0) + 1
	}
	return h.sort
end

def range(l, k)
	
		l1 = l.map { |x| x[k]}
		a, sd, mx, mn = stat(l1)
		h = hist(l1)
		x = h.max_by { |k, v| v }
		return {"#{k}_a" => a, "#{k}_sd" => sd,
			"#{k}_mx" => mx, "#{k}_mn" => mn,
			"#{k}_md" => x[0]}
	
end

def stat2(l, t, n)
	return Hash[[["a#{n}", "sd#{n}", "mx#{n}"], stat(l)[0..2].map { |x| x / t }].transpose]
end

def d(s)
        c = s.split('').select { |x| x == '1' }.size
	d = c.to_f / s.length
        return d
        
end

def data(x)
	n = x['n']
	ns = n.to_s(2)
	nl = ns.length
	m = nl / 2
	
	nsh = ns[0..m]
	nsl = ns[m..-1]
	
	asdm1 = stat2(ns.split(/0+/).map { |x| x.length }, nl, 1)
	
	l1 = ns.split(/1+/)
	
	l1.shift
	asdm0 = stat2(l1.map { |x| x.length }, nl, 0)
	
	return {'n' => x['n'], 'ns' => x['ns'], 'h2' => x['h2'], 'd' => d(ns), 'dh' => d(nsh), 'dl' => d(nsl)}.merge(asdm1).merge(asdm0)
end

def fit(l, y, lx)

	a = {}

	(lx + [y]).each { |x| a[x] = l.map { |b| b[x] }.to_vector() }

	ds = a.to_dataset()

	r = Statsample::Regression.multiple(ds, y)
#	$stderr.puts(r.summary)

	return r.coeffs.merge({'c' => r.constant})

end

def predict(l, y, z)
	l.each \
	{
		|x|
		t = z['c']
		
		(z.keys - ['c']).each { |k| t += z[k] * x[k] }
		x["#{y}_p"] = t
	}
end

def solve(l, y)
	z = fit(l, 'h2', ['d', 'dh', 'dl', 'a0', 'sd0', 'mx0', 'a1', 'sd1', 'mx1'])
	predict(l, 'h2', z)
end

def out1(fn, l)
	
	f = File.open(fn, 'w')
	f.puts(l[0].keys.join("\t"))
	l.each { |x| f.puts(x.values.join("\t")) }
	f.close
end

def balance(l, f1, f2)

	h = {}
	l.each \
	{
		|x|
		h[x['h2']] = [] if (!h.member?(x['h2']))
		h[x['h2']] << x
	}
	l2 = []
	c = 10
	f = {'n' => f1, 'd' => f2}
	
	h.sort.select { |k, v| v.size >= c }.each \
	{
		|k, v|
		
		['n', 'd'].each \
		{
			|k|
			next if (!f[k])
			
			v.sort_by! { |x| x[k] }
			
			c.times \
			{
				|i|
				j = ((i.to_f / (c - 1)) * (v.size - 1)).to_i
				next if (v[j].member?('.'))
				v[j]['.'] = nil
				l2 << v[j]
			}
#			c.times { break if (v.empty?); l2 << v.delete_at(rand(v.size)) }
		}
	}
	
	return l2
end

def balance2(l)

	h = {}
	c = 10
	l.each \
	{
		|x|
		h[x['h2']] = [] if (!h.member?(x['h2']))
		h[x['h2']] << x if (h[x['h2']].size < c)
	}
	l2 = []
	h.sort.select { |k, v| v.size == c }.each \
	{
		|k, v|
		l2 += v
	}
	
	return l2
end

def sum(l)

	t = 0.0
	l.each { |x| t += x }
	return t
end

def av(l)
	return nil if (l.empty?)
	return sum(l) / l.size
end

def coef(l)
	l = l.map { |x| data(x) }
	begin
		solve(l, 'h2')
	rescue Statsample::Regression::LinearDependency
		return nil
	end
	
	xav = av(l.map { |x| x['h2'] })
	yav = av(l.map { |x| x['h2_p'] })
	
	tx = ty = txy = e = 0.0
	l.each \
	{
		|z|
		
		x = z['h2']
		y = z['h2_p']
		txy += (x - xav) * (y - yav)
		tx += (x - xav) ** 2
		ty += (y - yav) ** 2
		e += (x - y) ** 2
	}
	r = txy / (Math.sqrt(tx) * Math.sqrt(ty))
	e /= l.size
	return r, e
end

def tobool(x)
	
	return (0..2).map { |y| x[y, 1] == '1' }
	
end

def out(fn, a)
	
	f = File.open(fn, 'a')
	f.puts(a.keys.join("\t")) if (f.size == 0)
	f.puts(a.values.join("\t"))
	f.close
end

def avg(a, k, x)
	return if (x.nil?)
	a[k] = {'t' => 0.0, 'c' => 0.0} if (!a.member?(k))
	a[k]['t'] += x
	a[k]['c'] += 1
	a[k]['a'] = a[k]['t'] / a[k]['c']
end

fn = 'out.txt'

File.open(fn, 'w').close


w = 20

a = {}
loop \
{
	l = dist(w, [true, true, true])
	l = balance2(l)
	r, e = coef(l)
	out(fn, {'r' => r, 'e' => e, 'w' => w, 'c' => l.size}) if (!r.nil?)
	w += 1
}