MT3. 拜访
描述
现在有一个城市销售经理,需要从公司出发,去拜访市内的某位商家,已知他的位置以及商家的位置,但是由于城市道路交通的原因,他每次移动只能在左右中选择一个方向 或 在上下中选择一个方向,现在问他有多少种最短方案到达商家地址。
示例1
输入:
[[0,1,0],[2,0,0]],2,3
输出:
2
示例2
输入:
[[2,0,0,0],[0,-1,-1,0],[0,-1,1,0],[0,0,0,0]],4,4
输出:
2
C++ 解法, 执行用时: 3ms, 内存消耗: 412KB, 提交时间: 2022-08-06
class Solution { public: /** * 代码中的类名、方法名、参数名已经指定,请勿修改,直接返回方法规定的值即可 * * * @param CityMap int整型vector<vector<>> * @param n int整型 * @param m int整型 * @return int整型 */ int dx[4] = {1, 0, -1, 0}; int dy[4] = {0, 1, 0, -1}; int countPath(vector<vector<int> >& CityMap, int n, int m) { vector<int> d(n * m, 0x3f3f3f3f), f(n * m, 0); int src, dst; for(int i = 0; i < n ; i++){ for(int j = 0; j < m; j++){ if(CityMap[i][j] == 1){ src = i * m + j; }else if(CityMap[i][j] == 2){ dst = i * m + j; } } } queue<int> q; q.push(src); d[src] = 0, f[src] = 1; while(!q.empty()){ int t = q.front(); q.pop(); int x = t / m, y = t % m, nx, ny, nt; for(int i = 0; i < 4; i++){ nx = x + dx[i], ny = y + dy[i]; nt = nx * m + ny; if(nx >= 0 && nx < n && ny >= 0 && ny < m && CityMap[nx][ny] != -1){ if(d[nt] > d[t] + 1){ d[nt] = d[t] + 1; q.push(nt); } if(d[nt] == d[t] + 1) f[nt] += f[t]; } } } return f[dst]; } };
C++ 解法, 执行用时: 3ms, 内存消耗: 424KB, 提交时间: 2022-01-28
class Solution { public: /* 方法一:深度优先搜索 复杂度分析 时间复杂度:需要遍历地图上所有的位置,所以时间复杂度是O(n∗m) 空间复杂度:O(n*m) */ int countPath1(vector<vector<int> >& CityMap, int n, int m) { row = n; col = m; visited.resize(n); for(int i = 0; i < n; ++i) { visited[i].resize(m); } for(int i = 0; i < n; ++i) { for(int j = 0; j < m; ++j) { if(CityMap[i][j] == 2) { startx = i; starty = j; } } } dfs(CityMap, startx, starty); return count; } /* 方法二:广度优先搜索+动态规划 */ int countPath(vector<vector<int> >& CityMap, int n, int m) { //path_nums为dp数组,用于记录起点到当前位置最短距离的方案数 int visited[n][m], path_nums[n][m]; int directions[4][2] = {{0, 1}, {0, -1}, {-1, 0}, {1, 0}}; memset(visited, 0, sizeof(visited)); queue<pair<int, int>> Que; for(int i = 0; i < n; ++i) for(int j = 0; j < m; ++j) if(CityMap[i][j] == 1) { Que.push({i, j}); path_nums[i][j] = 1; visited[i][j] = 1; } while(!Que.empty()) { int size = Que.size(); while(size--) { auto t = Que.front(); Que.pop(); for(int i = 0; i < 4; ++i) { int new_x = t.first + directions[i][0], new_y = t.second + directions[i][1]; if(new_x < 0 || new_x >= n || new_y < 0 || new_y >= m || CityMap[new_x][new_y] == -1) continue; if(!visited[new_x][new_y]) { visited[new_x][new_y] = visited[t.first][t.second] + 1; path_nums[new_x][new_y] = path_nums[t.first][t.second]; Que.push({new_x, new_y}); } else { if(visited[new_x][new_y] == visited[t.first][t.second] + 1) path_nums[new_x][new_y] += path_nums[t.first][t.second]; } } } } for(int i = 0; i < n; ++i) for(int j = 0; j < m; ++j) if(CityMap[i][j] == 2) return path_nums[i][j]; return 0; } private: int dires[4][2] = {{1,0}, {-1,0}, {0,1}, {0,-1}}; vector<vector<int>> visited; int pathlen = 0, shortestlen = 101; int count = 0; int row, col; int startx, starty; void dfs(vector<vector<int> >& CityMap, int x, int y) { int n = row, m = col; //路径长度大于最优即返回,减少计算次数 //如果遇到越界、遇到障碍物、已经访问过、或者距离更大的情况,直接返回 if(x<0 || x>=n || y<0 || y>=m || CityMap[x][y]==-1 || visited[x][y]==1 || pathlen>=shortestlen){ return; } ++pathlen; visited[x][y] = 1; if(CityMap[x][y] == 1){ if(pathlen < shortestlen){ shortestlen = pathlen; count = 1; } else if(pathlen == shortestlen){ //找到一条合法路径 ++count; } visited[x][y] = 0; --pathlen; return; } //往四个方向进行深度优先搜索 for(int i = 0; i < 4; ++i){ int xx = x + dires[i][0]; int yy = y + dires[i][1]; dfs(CityMap, xx, yy); } visited[x][y] = 0; //每次搜索完,重置标记数组 --pathlen; } };
C++ 解法, 执行用时: 3ms, 内存消耗: 432KB, 提交时间: 2022-07-24
class Solution { public: /** * 代码中的类名、方法名、参数名已经指定,请勿修改,直接返回方法规定的值即可 * * * @param CityMap int整型vector<vector<>> * @param n int整型 * @param m int整型 * @return int整型 */ vector<int> dx = {1,-1,0,0}; vector<int> dy = {0,0,1,-1}; int countPath(vector<vector<int> >& CityMap, int n, int m) { // write code here int res = 0; vector<vector<int> > vis(n,vector<int>(m,0)); queue<int> q; for(int i=0;i<n;i++){ for(int j=0;j<m;j++){ if(CityMap[i][j]==1) { q.emplace(i*m+j); q.emplace(-1); vis[i][j]=1; break; } } if(!q.empty()) break; } bool flag = false; while(!q.empty()){ if(q.front()==-1 && flag){ break; } if(q.front()==-1){ q.pop(); q.emplace(-1); continue; } int x = q.front()/m, y = q.front()%m; vis[x][y] = 1; q.pop(); if(CityMap[x][y] == 2){ flag=true; res++; continue; } if(flag) continue; for(int i=0;i<4;i++){ int xx = x+dx[i], yy = y+dy[i]; if(xx<0 || xx>=n || yy<0 || yy>=m || CityMap[xx][yy] == -1 || vis[xx][yy] == 1) continue; q.emplace(xx*m+yy); } } return res; } };
C++ 解法, 执行用时: 3ms, 内存消耗: 468KB, 提交时间: 2022-05-21
class MPoint{ public: int m_x; int m_y; MPoint(int x, int y):m_x(x),m_y(y){ } }; class Solution { public: /** * 代码中的类名、方法名、参数名已经指定,请勿修改,直接返回方法规定的值即可 * * * @param CityMap int整型vector<vector<>> * @param n int整型 * @param m int整型 * @return int整型 */ bool isvaild(vector<vector<int> >& CityMap, int x, int y){ int n = CityMap.size(); int m = CityMap[0].size(); if(x<0 || x>=m || y<0 || y>=n || CityMap[y][x] == -1) return false; return true; } MPoint next_points(MPoint point,int i){ const int direction[4][2] = {0,1,0,-1,1,0,-1,0}; return MPoint(point.m_x + direction[i][0], point.m_y + direction[i][1]); } int countPath(vector<vector<int> >& CityMap, int n, int m) { // write code here queue<MPoint> q; int start_x, start_y; int end_x, end_y; for(int i=0;i<n;i++){ for(int j=0;j<m;j++){ if(CityMap[i][j] == 2){ end_x = j; end_y = i; continue; } if(CityMap[i][j] == 1){ start_x = j; start_y = i; continue; } } } vector<vector<int>> visited(n,vector<int>(m, 0)); vector<vector<int>> distance(n,vector<int>(m,0)); vector<vector<int>> path_counts(n,vector<int>(m, 0)); q.push(MPoint(start_x, start_y)); visited[start_y][start_x] = 1; path_counts[start_y][start_x] = 1; int step = 0; int count = 0; while(!q.empty()){ int sz = q.size(); while(sz--){ auto point = q.front(); q.pop(); if(CityMap[point.m_y][point.m_x] == 2) return path_counts[point.m_y][point.m_x]; for(int i=0;i<4;i++){ auto next_point = next_points(point, i); if(isvaild(CityMap, next_point.m_x, next_point.m_y)){ path_counts[next_point.m_y][next_point.m_x] += path_counts[point.m_y][point.m_x]; if(visited[next_point.m_y][next_point.m_x] ==0) q.push(next_point); visited[next_point.m_y][next_point.m_x] = 1; } } } } cout << "fdsa"; return path_counts[end_y][end_x];; } };
C++ 解法, 执行用时: 3ms, 内存消耗: 488KB, 提交时间: 2022-03-15
class Solution { public: /** * 代码中的类名、方法名、参数名已经指定,请勿修改,直接返回方法规定的值即可 * * * @param CityMap int整型vector<vector<>> * @param n int整型 * @param m int整型 * @return int整型 */ int minCost = 999; //记录最短路径的距离 int cnt = 0; //记录最短路径的方案 int cost = 0; //记录路径距离 int dx[4] = {0, 1, 0, -1}; //右 下 左 上 int dy[4] = {1, 0, -1, 0}; int ex, ey, sx, sy; vector<vector<bool>> vis; void dfs(vector<vector<int>>& CityMap, int sx, int sy){ int n = CityMap.size(), m = CityMap[0].size(); //220315_DXM_这比放在for循环中调用时间短 if(sx < 0 || sx >= n || sy < 0 || sy >= m || CityMap[sx][sy] == -1 || vis[sx][sy] == true || cost >= minCost) return; //220315_DXM_可访问,设置为已访问状态 cost++; vis[sx][sy] = true; if(CityMap[sx][sy] == 2){ if(cost < minCost){ minCost = cost; cnt = 1; //重置最短路径 } else if(cost == minCost) cnt++; //220315_DXM_回溯 vis[sx][sy] = false; cost--; return; } for(int i = 0; i < 4; i++){ int nx = sx + dx[i], ny = sy + dy[i]; dfs(CityMap, nx, ny); } vis[sx][sy] = false; cost--; } int countPath(vector<vector<int> >& CityMap, int n, int m) { int ans = 0; vis.resize(n); for(int i = 0; i < n; i++) vis[i].resize(m); //220315_DXM_找到地图中的起点和终点 for(int i = 0; i < n; i++){ for(int j = 0; j < m; j++){ if(CityMap[i][j] == 1) sx = i, sy = j; } } dfs(CityMap, sx, sy); return cnt; } };