leetcode1066

 1 import sys
 2 class Solution:
 3     def __init__(self):
 4         self.sums = sys.maxsize
 5 
 6     def backtrack(self,workers,i,bikes,used,sums):
 7         if i == len(workers):
 8             self.sums = min(self.sums,sums)
 9             return
10 
11         if sums > self.sums:
12             return
13 
14         for j in range(len(bikes)):
15             if used[j]:
16                 continue
17             used[j] = True
18             self.backtrack(workers,i+1,bikes,used,sums+self.getDist(workers[i],bikes[j]))
19             used[j] = False
20 
21     def getDist(self,worker,bike):
22         return  abs(worker[0]-bike[0]) + abs(worker[1]-bike[1])
23 
24     def assignBikes(self, workers: 'List[List[int]]', bikes: 'List[List[int]]') -> int:
25         v = [False] * len(bikes)
26         self.backtrack(workers,0,bikes,v,0)
27         return self.sums

上面的代码使用回溯法，但是TLE。考虑需要使用动态规划提升效率，目前还做不出来。

 

看到一个同样思路的解决方案，使用java语言，却可以AC，这难道又是leetcode平台的“语言杀”？

这种同样思路，使用java可以AC，但是使用.net或python却TLE的情况，我是遇到过几次的。

但也有可能是我的代码写的有问题，目前还没弄明白具体原因。

java版本的代码如下：

 1 class Solution {
 2     public int assignBikes(int[][] workers, int[][] bikes) {
 3         dfs(workers, 0, bikes, new boolean[bikes.length], 0);
 4         return min;
 5     }
 6     
 7     int min = Integer.MAX_VALUE;
 8     void dfs(int[][] workers, int i, int[][] bikes, boolean[] used, int sum) {
 9         if (i == workers.length) {
10             min = Math.min(min, sum);
11             return;
12         }
13         
14         if (sum > min) return;  // early termination
15         
16         for (int j = 0; j < bikes.length; ++j) {
17             if (used[j]) continue;
18             used[j] = true;
19             dfs(workers, i+1, bikes, used, sum + getDistance(workers[i], bikes[j]));
20             used[j] = false;
21         }
22     }
23     
24     int getDistance(int[] worker, int[] bike) {
25         return Math.abs(worker[0] - bike[0]) + Math.abs(worker[1] - bike[1]);
26     }
27 }

参考地址：https://leetcode.com/problems/campus-bikes-ii/discuss/303375/Java-straightforward-DFS-solution

转载于:https://www.cnblogs.com/asenyang/p/10961846.html