随机一个从n到m的数，如何让数越大随机到的概率就越低。

我为鱼肉

随机一个从n到m的数，如何让数越大随机到的概率就越低。



已经解决，这是我按照菜刀提供的思路，定义的方法（比用数组或许能快一点点吧）。

RUBY 代码复制

module Kernel
  def random(range)
    first = range.first
    last = range.exclude_end? ? range.last - 1 : range.last
    max = (last - first + 1) * (first + last) / 2    
    result = rand(max) + 1
    for i in range
      small = i > first ? ((i - first) * (first + i - 1) / 2) : 0
      large = i > first ? ((i - first + 1) * (first + i) / 2) : first
      break if result > small and result <= large
      #return i if result > small and result <= large
    end
    return last - i + first
  end
end
#代码测试
a = []
1000.times{
r = random(20..40)
if a[r].nil?
  a[r] = 0
end
a[r] += 1
}
a.delete(nil)
p a
exit

module Kernel


  def random(range)


    first = range.first


    last = range.exclude_end? ? range.last - 1 : range.last


    max = (last - first + 1) * (first + last) / 2    


    result = rand(max) + 1


    for i in range


      small = i > first ? ((i - first) * (first + i - 1) / 2) : 0


      large = i > first ? ((i - first + 1) * (first + i) / 2) : first


      break if result > small and result <= large


      #return i if result > small and result <= large


    end


    return last - i + first


  end


end


#代码测试


a = []


1000.times{


r = random(20..40)


if a[r].nil?


  a[r] = 0


end


a[r] += 1


}


a.delete(nil)


p a


exit

灯笼菜刀王

f = []
max = m + 1
for i in n..m
  (max - i).times{f.push(i)}
end
p f[rand(f.size)]

guoxiaomi

使用非线性映射就行：y = (rand(x) ** 0.5 * (M - N) + N).to_i，只要函数相对y = x上凸就行