[英]Time Complexity of permutation function
Given a collection of distinct numbers, return all possible permutations.给定一组不同的数字,返回所有可能的排列。
For example, [1,2,3] have the following permutations:例如,[1,2,3] 有以下排列:
[ [1,2,3], [1,3,2], [2,1,3], [2,3,1], [3,1,2], [3,2,1] ] [ [1,2,3], [1,3,2], [2,1,3], [2,3,1], [3,1,2], [3,2,1] ]
My Iterative Solution is:我的迭代解决方案是:
public List<List<Integer>> permute(int[] nums) {
List<List<Integer>> result = new ArrayList<>();
result.add(new ArrayList<>());
for(int i=0;i<nums.length;i++)
{
List<List<Integer>> temp = new ArrayList<>();
for(List<Integer> a: result)
{
for(int j=0; j<=a.size();j++)
{
a.add(j,nums[i]);
List<Integer> current = new ArrayList<>(a);
temp.add(current);
a.remove(j);
}
}
result = new ArrayList<>(temp);
}
return result;
}
My Recursive Solution is:我的递归解决方案是:
public List<List<Integer>> permuteRec(int[] nums) {
List<List<Integer>> result = new ArrayList<List<Integer>>();
if (nums == null || nums.length == 0) {
return result;
}
makePermutations(nums, result, 0);
return result;
}
void makePermutations(int[] nums, List<List<Integer>> result, int start) {
if (start >= nums.length) {
List<Integer> temp = convertArrayToList(nums);
result.add(temp);
}
for (int i = start; i < nums.length; i++) {
swap(nums, start, i);
makePermutations(nums, result, start + 1);
swap(nums, start, i);
}
}
private ArrayList<Integer> convertArrayToList(int[] num) {
ArrayList<Integer> item = new ArrayList<Integer>();
for (int h = 0; h < num.length; h++) {
item.add(num[h]);
}
return item;
}
According to me the time complexity(big-Oh) of my iterative solution is: n * n(n+1)/2~ O(n^3)根据我的说法,我的迭代解决方案的时间复杂度(大哦)是: n * n(n+1)/2~ O(n^3)
I am not able to figure out the time complexity of my recursive solution.我无法计算出我的递归解决方案的时间复杂度。
Can anyone explain complexity of both?谁能解释两者的复杂性?
The recursive solution has a complexity of O(n!)
as it is governed by the equation: T(n) = n * T(n-1) + O(1)
.递归解决方案的复杂度为O(n!)
因为它由以下方程控制: T(n) = n * T(n-1) + O(1)
。
The iterative solution has three nested loops and hence has a complexity of O(n^3)
.迭代解决方案具有三个嵌套循环,因此具有O(n^3)
的复杂性。
However, the iterative solution will not produce correct permutations for any number apart from 3
.但是,迭代解决方案不会为除3
之外的任何数字生成正确的排列。
For n = 3
, you can see that n * (n - 1) * (n-2) = n!
对于n = 3
,您可以看到n * (n - 1) * (n-2) = n!
. . The LHS is O(n^3)
(or rather O(n^n)
since n=3
here) and the RHS is O(n!)
. LHS 是O(n^3)
(或者更确切地说是O(n^n)
因为这里n=3
),RHS 是O(n!)
。
For larger values of the size of the list, say n
, you could have n
nested loops and that will provide valid permutations.对于列表大小的较大值,例如n
,您可以有n
嵌套循环,这将提供有效的排列。 The complexity in that case will be O(n^n)
, and that is much larger than O(n!)
, or rather, n! < n^n
在这种情况下,复杂性将是O(n^n)
,这比O(n!)
大得多,或者更确切地说,是n! < n^n
n! < n^n
. n! < n^n
。 There is a rather nice relation called Stirling's approximation which explains this relation.有一个相当不错的关系称为斯特林近似,它解释了这种关系。
It's the output (which is huge) matters in this problem, not the routine's implementation.在这个问题中,输出(这是巨大的)很重要,而不是例程的实现。 For n
distinct items, there are n!
对于n
不同的项目,有n!
permutations to be returned as the answer, and thus we have at least O(n!)
complexity.排列作为答案返回,因此我们至少有O(n!)
复杂度。
With a help of Stirling's approximation借助斯特林近似
O(n!) = O(n^(1/2+n)/exp(n)) = O(sqrt(n) * (n/e)^n)
we can easily see, that O(n!) > O(n^c)
for any constant c
, that's why it doesn't matter if the implementation itself adds another O(n^3)
since我们可以很容易地看到,对于任何常量c
, O(n!) > O(n^c)
,这就是为什么实现本身添加另一个O(n^3)
并不重要,因为
O(n!) + O(n^3) = O(n!)
In terms of number of times the method makePermutations
is called, the exact time complexity would be:就调用makePermutations
方法的次数而言,确切的时间复杂度为:
O( 1 + n + n(n-1) + n(n-1)(n-2) + ... )
For n = 3:对于 n = 3:
O( 1 + 3 + (3*2) + (3*2*1) ) = O(16)
This means, for n = 3, the method makePermutations
will be called 16 times.这意味着,对于 n = 3,方法makePermutations
将被调用 16 次。
And I think the space complexity for an optimal permutations function would be O(n * n,) because there are n.我认为最佳排列 function 的空间复杂度为 O(n * n,),因为有 n。 total arrays to return, and each of those arrays is of size n.返回的总数为 arrays,其中每个 arrays 的大小为 n。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.