break algorithm---暴力搜索:回溯(1)

break algorithm---暴力搜索:回溯(1),第1张


😡😡😡

声明:
算法基于https://labuladong.github.io/
python语言实现


core mind
46.permutations
47.permutations-ii
51.n-queens
77.combinations
78.subsets
90.subsets-ii
698.partition-to-k-equal-sum-subsets


😡😡😡

core mind

【回溯算法秒杀所有排列/组合/子集问题】https://labuladong.github.io/algo/4/29/107/


"""
【回溯问题】<=>【决策树遍历】<=>【纯暴力穷举】<=>【无重复子问题的递归】:4 key point[choose list + choose + track + end state]

解决一个回溯问题,实际上就是一个决策树的遍历过程。

站在回溯树的一个节点上,你只需要思考 3 个问题:
1、【选择列表】:也就是你当前可以做的选择 => 对应题目中给出的输入列表
2、【选择】:选择列表指针i,用于遍历选择列表,作出选择 ,递归过程:
    def backtrack(i:int[输入列表指针], track:List[存储符合题意的选择,即路径列表])
        for i in range(len(input_list)):
            # 2.1 递归出口:选择列表指针i移动至表尾,所有元素均得到处理后,使用return语句返回
            if i==len(input_list):
                1.if [satisfy answer]:return True
                2.or res.append(track[.copy()])
            # 2.2 核心递归部分:选择列表指针i逐步移动至表尾
                => 不符合题意的选择:if [choose unsatisfied]: continue
                                continue语句跳过该选择,进入下一次循环
                => 符合题意的选择:if [choose satisfied]:
                                a 选择列表--移除--该选择: input_list.remove(input_list[i])
                                b 路径列表--加入--该选择: track.append(input_list[i])
                [if] backtrack(i+1, track) [return]  # 将选择列表指针i后移这一动作打入递归调用过程
                => 递归调用后撤销【符合题意的选择】,【还原决策树】
                                a 选择列表--加入--该选择: input_list.append(input_list[i])
                                b 路径列表--移除--该选择: track.remove(input_list[i])

3、【路径】:存储于已经做出的、符合题意的选择。
4、【结束条件】:也就是到达决策树底层,无法再做选择的条件。


"""


# 核心就是 for 循环里面的递归,在递归调用之前「做选择」,在递归调用之后「撤销选择」
'''
class Solution:
    def outer_func(self, nums: List[int]) -> [return_type]:
    # 记录结果集:存储各个路径列表track
    result = []
    # 记录符合题意的选择列表指针i对应的值nums[i]
    track = []
    
    # 仅限排列使用:标记数组used,用于记录nums[i]是否被使用
    used = [False for i in range(len(nums))] 
    
    组合|子集:def backtrack(选择列表指针i, 存储符合题意的选择nums[i]路径列表track):
    排列:def backtrack(存储符合题意的选择nums[i]路径列表track, 标记数组used):
        if 满足结束条件:
            result.add(track)
            return
    
        for 选择列表指针i in range(len(选择列表nums)):
            1. 选择判断:不符合 or 符合 题意
            1.1 针对不符合题意的选择:
                if [choose unsatisfied]: continue
                    continue语句跳过该选择,进入下一次循环
            1.2 针对符合题意的选择:记录在存储【符合题意的选择列表指针i对应的值】的 【路径列表track】中
                将该选择从选择列表移除:used[i]=True
                路径.add(选择):track.append(nums[i])
                
            2.递归调用:
            组合|子集:backtrack(选择列表指针i后移一位i+1, 路径列表track) 
                                # 通过保证元素之间的相对顺序不变【即在递归调用过程中使 列表指针变量i+1】来防止出现重复的子集。
            排列:backtrack(路径列表track, 记录指针变量i是佛被访问过的used标记数组)
            
            3.撤销选择:(递归后撤销选择,还原该决策树,进行下一次决策树递归过程)将符合题意的选择列表指针值 从 路径列表track中移除
                路径列表.remove(选择):track.remove(nums[i])
                将该选择再加入选择列表:used[i]=False
'''


😡😡😡

46.permutations
'''
NOTE:LeetCode提示超时时,去掉导包语句及所有注释即可通过
'''
from typing import List


class Solution:
    '''
    【排列(元素无重不可复选)】
    '''
    def permute(self, nums: List[int]) -> List[List[int]]:
        # 最终结果集result:存储track
        res = []

        # 符合题意的选择列表指针集:存储符合题意的选择列表指针i对应的值nums[i]
        track = []
        # 标记数组used,用于记录nums[i]是否被使用
        used = [False for j in range(len(nums))]

        def backtrack(track: List[int], used: List[bool]):
            # 结束条件:当track的长度等于nums的长度时,说明已经完成了一个全排列
            if len(track) == len(nums):
                # NOTE:此处添加的是track的副本,否则当后续撤销选择时,track指向的内容将不存在,导致res全为空
                res.append(track.copy())
                return

            for i in range(len(nums)):
                # 1.1 针对不符合题意的选择i:跳过该选择,进行下一次循环判断
                if used[i]: continue

                # 1.2 针对符合题意的选择i:加入存储 符合题意的选择 的路径列表tarck ,并将i标记为已经被使用过。
                track.append(nums[i])
                used[i] = True

                # 2. core:recursive
                backtrack(track, used)

                # 3.递归之后撤销选择,还原该决策树:路径列表中移出nums[i]这一选择,并将i标记为未被使用。
                track.remove(nums[i])
                used[i] = False

        backtrack(track, used)
        return res


def main():
    # Input: nums = [1,2,3]
    # Output: [[1,2,3],[1,3,2],[2,1,3],[2,3,1],[3,1,2],[3,2,1]]
    solution1 = Solution()
    print(solution1.permute(nums=[1, 2, 3]))

    # Input: nums = [0,1]
    # Output: [[0,1],[1,0]]
    solution2 = Solution()
    print(solution2.permute(nums=[0, 1]))

    # Input: nums = [1]
    # Output: [[1]]
    solution3 = Solution()
    print(solution3.permute(nums=[1]))


if __name__ == '__main__':
    main()


😡😡😡

47.permutations-ii
'''
solution3:存在问题,带解决
'''
from typing import List


class Solution:
    '''
    【排列(元素可重不可复选)】
    '''

    def permuteUnique(self, nums: List[int]) -> List[List[int]]:
        res = []
        track = []
        used = [False for i in range(len(nums))]
        # nums.sort()

        def backtrack(nums: List[int]):
            # 递归出口
            if len(track) == len(nums):
                res.append(track.copy())
                return

            for i in range(len(nums)):
                if i > 0 and nums[i] == nums[i - 1] and nums[i - 1] is False:
                    continue

                if used[i] == True:
                    continue
                elif used[i] == False:
                    track.append(nums[i])
                    used[i] = True

                backtrack(nums)

                track.remove(nums[i])
                used[i] = False

        backtrack(nums)
        # res_sorted = [sorted(i) for i in res]
        res_nodup_tuple = list(set([tuple(i) for i in res]))
        res_nodup_list = [list(i) for i in res_nodup_tuple]
        return res_nodup_list


def main():
    # Input: nums = [1,1,2]
    # Output:
    # [[1,1,2],
    #  [1,2,1],
    #  [2,1,1]]
    solution1 = Solution()
    print(solution1.permuteUnique(nums=[1, 1, 2]))

    # Input: nums = [1,2,3]
    # Output: [[1,2,3],[1,3,2],[2,1,3],[2,3,1],[3,1,2],[3,2,1]]
    solution2 = Solution()
    print(solution2.permuteUnique(nums=[1, 2, 3]))

    # Input:
    # [2,2,1,1]
    # Output:
    # [[1,2,1,2],[1,2,2,1],[2,2,1,1],[2,1,1,2],[1,1,2,2]]
    # Expected:
    # [[1,1,2,2],[1,2,1,2],[1,2,2,1],[2,1,1,2],[2,1,2,1],[2,2,1,1]]
    solution3 = Solution()
    print(solution3.permuteUnique([2, 2, 1, 1]))


if __name__ == '__main__':
    main()


😡😡😡

51.n-queens
from typing import List


class Solution:
    def solveNQueens(self, n: int) -> List[List[str]]:
        # 最终结果集result:存储board
        res = []

        # 符合题意的选择列表指针集:存储符合题意的选择列表指针row 对应的值
        # board存储的数据结构[".Q..","...Q","Q...","..Q."]
        board = [['.'] * n for i in range(n)]

        def isValid(board: List, row: int, column: int) -> bool:
            # 1.该格所在的整列不能有Q
            for i in range(n):
                if board[i][column] == 'Q':
                    return False
            # 2.该格左上连线不能有Q
            i = row - 1
            j = column - 1
            while i >= 0 and j >= 0:  # NOTE:边界条件取值范围,是否带等
                if board[i][j] == 'Q':
                    return False
                else:
                    i -= 1
                    j -= 1
            # 3.该格右上连线不能有Q
            i = row - 1
            j = column + 1
            while i >= 0 and j <= n - 1:
                if board[i][j] == 'Q':
                    return False
                else:
                    i -= 1
                    j += 1
            return True

        def backtrack(board: List, row: int):
            # 1.回溯出口:找到一个满足条件的board
            if row == len(board):
                board_str = [''.join(list_i) for list_i in board]
                res.append(board_str)
                return

            for column in range(n):
                # 1. 选择判断:不符合 or 符合 题意
                # 1.1 针对不符合题意的选择:
                if isValid(board, row, column) == False:
                    continue
                # 1.2 针对符合题意的选择:记录在board中
                board[row][column] = 'Q'

                # 2. 核心递归过程:进入下一层决策树,即在下一行放皇后Q
                backtrack(board, row + 1)

                # 3.递归后:撤销选择,将针对符合题意的选择从board中移除
                board[row][column] = '.'

        # 选择列表 行指针row 从0向下一次移动
        backtrack(board, 0)
        return res


def main():
    # Input: n = 4
    # Output: [[".Q..","...Q","Q...","..Q."],["..Q.","Q...","...Q",".Q.."]]
    # Explanation: There exist two distinct solutions to the 4-queens puzzle as shown above
    solution1 = Solution()
    print(solution1.solveNQueens(n=4))

    # Input: n = 1
    # Output: [["Q"]]
    solution2 = Solution()
    print(solution2.solveNQueens(n=1))


if __name__ == '__main__':
    main()


😡😡😡

77.combinations
from typing import List


class Solution:
    '''
    【组合(元素无重不可复选)】
    '''
    def combine(self, n: int, k: int) -> List[List[int]]:
        # 结果集res:List[List[int]],存储各个track
        res = []
        # 符合题意的选择集
        track = []

        # 元素顺序保证start
        def backtrack(start):
            # 递归出口
            if len(track) == k:
                res.append(track.copy())
                return

                # NOTE:指针变量i每次从start位开始递归
            for i in range(start, n + 1, 1):
                track.append(i)
                backtrack(i + 1)
                track.remove(i)

        backtrack(1)
        return res


def main():
    # Input: n = 4, k = 2
    # Output:
    # [
    #   [2,4],
    #   [3,4],
    #   [2,3],
    #   [1,2],
    #   [1,3],
    #   [1,4],
    # ]
    solution1 = Solution()
    print(solution1.combine(n=4, k=2))

    print('---')

    # Input: n = 1, k = 1
    # Output: [[1]]
    solution2 = Solution()
    print(solution2.combine(n=1, k=1))


if __name__ == '__main__':
    main()


😡😡😡

78.subsets
from typing import List

class Solution:
    '''
    【子集(元素无重不可复选)】
    '''
    # Input: nums = [1,2,3]
    # Output: [[],[1],[2],[1,2],[3],[1,3],[2,3],[1,2,3]]
    # 子集问题 <==> 组合问题 :通过保证元素之间的相对顺序不变【即在递归调用过程中使:列表指针变量i+1】来防止出现重复的子集
    def subsets(self, nums: List[int]) -> List[List[int]]:
        # 结果集res:List[List[int]],存储各个track
        res = []
        # 记录各个子集:List[int]
        track = []

        def backtrack(start: int):
            # 0.结果集更新位置
            res.append(track.copy())

            # 2.注意i的起始位start
            # NOTE:以下语句将会时i在0、1之间反复横跳,导致无限递归
            # for i in range(len(nums)):
            for i in range(start, len(nums), 1):
                # 递归前:将选择i的值 加入 路径列表track
                track.append(nums[i])
                # 递归:通过保证元素之间的相对顺序不变【即在递归调用过程中使:列表指针变量i+1】来防止出现重复的子集
                backtrack(i + 1)
                # 递归后:将选择i的值 移出于 路径列表track
                track.remove(nums[i])

        backtrack(0)
        return res


def main():
    # Input: nums = [1,2,3]
    # Output: [[],[1],[2],[1,2],[3],[1,3],[2,3],[1,2,3]]
    solution1 = Solution()
    print(solution1.subsets(nums=[1, 2, 3]))

    print('---')

    # Input: nums = [0]
    # Output: [[],[0]]
    solution2 = Solution()
    print(solution2.subsets(nums=[0]))


if __name__ == '__main__':
    main()



😡😡😡

90.subsets-ii
from typing import List


class Solution:
    '''
    【子集|组合(元素可重不可复选)】
        需要进行剪枝,如果一个节点有多条值相同的树枝相邻,则只遍历第一条,剩下的都剪掉,不要去遍历。
        体现在代码上,需要先进行排序,让相同的元素靠在一起,如果发现 nums[i] == nums[i-1],则跳过。
    '''

    # Input: nums = [1,2,2]
    # Output: [[],[1],[1,2],[1,2,2],[2],[2,2]]
    def subsetsWithDup(self, nums: List[int]) -> List[List[int]]:
        # 按升序排序,使相同元素靠在一起,后续在递归过程中对于相同元素只处理一次,后续相同元素跳过。
        nums.sort()
        res = []
        track = []

        def backtrack(start: int):
            # update res position
            # NOTE:此处添加的是路径列表track的一份拷贝值,否则在后续移除元素过程中致使track为空,最后导致res中什么也没加进去
            res.append(track.copy())

            for i in range(start, len(nums), 1):
                # 只处理一次相同元素,注意移动指针前需进行边界条件限制:
                #                                       指针前移i-1,需先保证 i>start
                #                                       指针后移1+1,需先保证 i
                # wrong:
                # [[], [1], [1, 2], [1, 2, 2], [1, 2], [2], [2, 2], [2]]
                # if nums[i] < len(nums) - 1 and nums[i] == nums[i + 1]:
                #     continue
                # right:
                # 剪枝逻辑,值相同的相邻树枝,只遍历第一条
                # [[], [1], [1, 2], [1, 2, 2], [2], [2, 2]]
                if i > start and nums[i] == nums[i - 1]:
                    continue
                track.append(nums[i])
                backtrack(i + 1)
                track.remove(nums[i])

        backtrack(0)
        return res


def main():
    # Input: nums = [1,2,2]
    # Output: [[],[1],[1,2],[1,2,2],[2],[2,2]]
    solution1 = Solution()
    print(solution1.subsetsWithDup(nums=[1, 2, 2]))

    # Input: nums = [0]
    # Output: [[],[0]]
    solution2 = Solution()
    print(solution2.subsetsWithDup(nums=[0]))


if __name__ == '__main__':
    main()


😡😡😡

698.partition-to-k-equal-sum-subsets
'''
NOTE:存在问题,还未解决
38 / 159 test cases passed.
Status: Time Limit Exceeded
'''
from typing import List


class Solution:
    def canPartitionKSubsets(self, nums: List[int], k: int) -> bool:

        sum = 0
        for p in range(len(nums)): sum += nums[p]
        if sum % k != 0 or k > len(nums): return False
        target = sum / k

        buckets = [0 for p in range(k)]

        def backtrack(i: int, buckets: List) -> bool:
            # 1.递归出口:指针变量i遍历到len(nums),
            # NOTE:出口条件为决策完最后一个元素[len(nums)-1],i前移一位后指向len(nums)
            if i == len(nums):
                for bucket in buckets:
                    if bucket != target: return False
                # 若每个桶的和都=目标值target,则可均分为k等分
                return True

            # 2.遍历k个桶
            for j in range(k):
                # 2.1 递归前做选择,两种情况:nums[i] 可 | 不可 放入桶buckets[j]?
                if buckets[j] + nums[i] > target:
                    continue
                elif buckets[j] + nums[i] <= target:
                    buckets[j] += nums[i]
                # 2.2 递归:将遍历nums这一过程打入递归
                # NOTE:递归部分需要有返回值,否则该递归调用结果没有返回值,递归调用结果被舍弃,默认将None赋给res
                if backtrack(i + 1, buckets): return True
                # 2.3 递归后撤销选择
                buckets[j] -= nums[i]
                # NOTE:nums[i]装入哪个桶都不行
                return False

        return backtrack(0, buckets)


def main():
    # Input: nums = [4,3,2,3,5,2,1], k = 4
    # Output: true
    # Explanation: It is possible to divide it into 4 subsets (5), (1, 4), (2,3), (2,3) with equal sums.
    solution1 = Solution()
    print(solution1.canPartitionKSubsets(nums=[4, 3, 2, 3, 5, 2, 1], k=4))

    # Input: nums = [1,2,3,4], k = 3
    # Output: false
    solution2 = Solution()
    print(solution2.canPartitionKSubsets(nums=[1, 2, 3, 4], k=4))

    # Input: nums = [730,580,401,659,5524,405,1601,3,383,4391,4485,1024,1175,1100,2299,3908], k = 4
    # sum=28668  /  4 = 7167
    # Output: false
    # solution3 = Solution()
    # print(solution3.canPartitionKSubsets(nums=[730,580,401,659,5524,405,1601,3,383,4391,4485,1024,1175,1100,2299,3908], k=4))


if __name__ == '__main__':
    main()

欢迎分享,转载请注明来源:内存溢出

原文地址: http://outofmemory.cn/langs/791868.html

(0)
打赏 微信扫一扫 微信扫一扫 支付宝扫一扫 支付宝扫一扫
上一篇 2022-05-05
下一篇 2022-05-05

发表评论

登录后才能评论

评论列表(0条)

保存