[英]how to call a recursion call N number of times, given the number N?
I have an array of numbers: S= {4,5}
and I want to check if this group creates the sum = 13
.我有一个数字数组:
S= {4,5}
,我想检查这个组是否创建sum = 13
。
In this case, yes: 4 + 4 + 5 = 13
在这种情况下,是:
4 + 4 + 5 = 13
Another example: s={4,5}
, sum = 6
-> no另一个例子:
s={4,5}
, sum = 6
-> no
I wrote a recursive function to solve this:我写了一个递归函数来解决这个问题:
public static boolean isSumOf(int [] s,int n)
{
if(n == 0)
return true;
if(n < 0)
return false;
return isSumOf(s,n-s[0]) || isSumOf(s,n-s[1]);
}
But this function works only for 2 numbers in the array.但此函数仅适用于数组中的 2 个数字。 I need to write a recursive function that will deal with N numbers, like
{4,9,3}
or {3,2,1,7}
etc.我需要编写一个递归函数来处理 N 个数字,例如
{4,9,3}
或{3,2,1,7}
等。
I'm not sure how can I do this?我不确定我该怎么做? How can I call a recursion N times, according to the length of the array?
如何根据数组的长度调用递归 N 次? Or maybe I should change my algorithm completely?
或者我应该完全改变我的算法? Also - I'm not allowed to use loops.
另外 - 我不允许使用循环。
return isSumOf(s,n-s[0]) || isSumOf(s,n-s[1]);
You can generalize this with a loop like this:你可以用这样的循环来概括这一点:
for (int i = 0; i < s.length; ++i) {
if (isSumOf(s,n-s[i])) return true;
}
return false;
But, since you can't use loops, you can write the equivalent loop as another recursive method:但是,由于您不能使用循环,您可以将等效循环编写为另一种递归方法:
boolean withoutLoop(int [] s,int n, int i) {
if (i >= s.length) return false;
return isSumOf(s,n-s[i]) || recurse(s, n, i+1);
}
and then call it like so from your isSumOf
method:然后从你的
isSumOf
方法中像这样调用它:
public static boolean isSumOf(int [] s,int n)
{
if(n == 0)
return true;
if(n < 0)
return false;
return withoutLoop(s, n, 0); // Change here.
}
Or, if you want to write it more concisely:或者,如果你想写得更简洁:
return (n == 0) || (n < 0 && withoutLoop(s, n, 0));
Break the problem down:分解问题:
Then think of a way to express summing an array as recursive task;然后想办法来表达总结数组作为递归任务; eg Sum of elements 1 to N is element 1 + Sum of elements 2 to N.
例如,元素 1 到 N 的总和是元素 1 + 元素 2 到 N 的总和。
Finally, turn that idea / expression into code.最后,将这个想法/表达转化为代码。
For any recursion problem, use the template:对于任何递归问题,请使用模板:
ResultType recursiveMethod(params) {
if( /* this is the simplest case */ ) {
return answer for the simplest case
} else {
partialResult = solve part of the problem
resultForRest = recursiveMethod(rest of problem)
}
}
Particularly for list processing, this becomes:特别是对于列表处理,这变为:
if(list is empty) {
return solution for an empty list
} else {
r = call self recursively for tail of list
return solution for head of list combined with r
}
(Where "head" is the first item, and "tail" is the rest. Tail may be empty.) (其中“head”是第一项,“tail”是其余项。tail 可能为空。)
For your problem, the simplest case is an empty array:对于您的问题,最简单的情况是空数组:
if(s.length == 0) {
return n == 0;
}
For the else
, the "part of the problem" is s[0]
and the "rest of the problem" is s[1]
onwards.对于
else
,“问题的一部分”是s[0]
而“问题的其余部分”是s[1]
。
...
} else {
int head = s[0];
int[] tail = Arrays.copyOfRange(s,1,s.length-1);
return isSumOf(tail, n - head);
}
The code would be cleaner (and probably more efficient) if you used a List
instead of an array directly, because you could then use List.subList()
instead of copyOfRange()
.如果您直接使用
List
而不是数组,代码会更清晰(并且可能更高效),因为您可以使用List.subList()
而不是copyOfRange()
。
You could also pass the whole array each time, along with an extra parameter indicating how much of the array has already been accounted for.您还可以每次都传递整个数组,以及一个额外的参数,该参数指示已经考虑了多少数组。
This should work:这应该有效:
public static boolean isSumOf(int [] s,int n)
{
if(n == 0)
return true;
if(n < 0)
return false;
for (int x: s) {
if (isSumOf(s, n-x)) {
return true;
}
}
return false;
}
UPDATE:更新:
Oh!哦! no loop, only recursion, you will need an extra argument:
没有循环,只有递归,你需要一个额外的参数:
public static boolean isSumOf(int [] s,int n)
{
if(n == 0)
return true;
if(n < 0)
return false;
return isSum2(s, n, 0);
}
public static boolean isSum2(int [] s,int n,int i)
{
if (i >= s.length)
return false;
return isSumOf(s,n-s[i]) || isSum2(s,n,i+1);
}
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.