Python实现加权图最小成本路径查找程序
假设我们有一个名为edges的二维整数列表,它表示一个无向图。输入中的每一行都表示一条边[u, v, w],这意味着节点u和v连接,并且边的权重为w。该图包含从0到n-1的n个节点。
路径的成本在此定义为边数与路径中任何边的最大权重的乘积。我们必须找出从节点0到节点n-1的最小可能成本,如果不存在这样的路径,则将答案声明为-1。
So, if the input is like edges = [ [0, 2, 100], [1, 2, 200], [1, 3, 100], [2, 3, 300] ], then the output will be 600
为了解决这个问题,我们将遵循以下步骤:
graph := 新建一个映射
weights := 新建一个映射
max_weight := 0
N := 0
对于edges中的每个u, v, w,执行:
将v插入graph[u]的末尾
将u插入graph[v]的末尾
weights[u, v] := w
weights[v, u] := w
N := N, u + 1, v + 1中的最大值
max_weight := max_weight, w中的最大值
result := 无穷大
当max_weight >= 0时,执行:
d, weight := bfs(0, max_weight)
如果d >= 0,则
result := result, d * weight中的最小值
max_weight := weight - 1
否则,
终止循环
如果result < 无穷大,则返回result,否则返回-1
定义函数bfs()。这将接收root和weight_cap
target := N - 1
Q := 包含root, 0, 0的双端队列
visited := [False] * N
visited[0] := True
当Q不为空时,执行:
v, d, current_weight := 从Q删除最后一个元素
如果v等于N - 1,则
返回d, current_weight
对于graph[v]中的每个w,执行:
如果visited[w]不为零,则
继续下一次迭代
new_weight := weights[v, w]
如果new_weight <= weight_cap,则
visited[w] := True
将(w, d + 1, max(current_weight, new_weight))添加到Q的左侧
返回-1, -1
示例
让我们来看下面的实现以更好地理解:
from collections import defaultdict, deque class Solution: def solve(self, edges): graph = defaultdict(list) weights = {} max_weight = 0 N = 0 for u, v, w in edges: graph[u].append(v) graph[v].append(u) weights[u, v] = w weights[v, u] = w N = max(N, u + 1, v + 1) max_weight = max(max_weight, w) def bfs(root, weight_cap): target = N - 1 Q = deque([(root, 0, 0)]) visited = [False] * N visited[0] = True while Q: v, d, current_weight = Q.pop() if v == N - 1: return d, current_weight for w in graph[v]: if visited[w]: continue new_weight = weights[v, w] if new_weight <= weight_cap: visited[w] = True zQ.appendleft((w, d + 1, max(current_weight, new_weight))) return -1, -1 result = float("inf") while max_weight >= 0: d, weight = bfs(0, max_weight) if d >= 0: result = min(result, d * weight) max_weight = weight - 1 else: break return result if result < float("inf") else -1 ob = Solution() print (ob.solve( [ [0, 2, 100], [1, 2, 200], [1, 3, 100], [2, 3, 300] ]))
输入
[ [0, 2, 100], [1, 2, 200], [1, 3, 100], [2, 3, 300] ]
输出
600