繁体   English   中英

为什么以下算法的时间复杂度(循环排序?!)是O(n)?

[英]Why time complexity of following algorithm(cycle sort?!) is O(n)?

循环排序的时间复杂度是O(n ^ 2) 参考

但是,该解决方案声称以下算法涉及循环排序仅使用O(n)。 时间复杂度不应该是O(n ^ 2)吗?

    def find_all_missing_numbers(nums):
        """
        :type nums: List[int]
        :rtype: List[int]
        """
        i = 0

        # This is cycle sort and should use O(n^2)?!
        while i < len(nums):
            j = nums[i] - 1
            if nums[i] != nums[j]:
                nums[i], nums[j] = nums[j], nums[i]  # swap
            else:
                i += 1

        missingNumbers = []

        # O(n)
        for i in range(len(nums)):
            if nums[i] != i + 1:
                missingNumbers.append(i + 1)

        return missingNumbers
# 

时间复杂度= O(n ^ 2 + n)= O(n ^ 2)。 解决方案错了吗?

这不是循环排序,如果数组由范围[1, len(array)]的数字组成,则该算法旨在查找缺少的数字。

print(find_all_missing_numbers([5,4,3,2,1]))
print(find_all_missing_numbers([1,2,3,5,5]))
print(find_all_missing_numbers([2]))

[]
[4]
错误

该行假设存储的数字给出了正确的位置,仅当数字在上面所示的范围内时才有效。

j = nums[i] - 1

循环排序花费线性时间为每个数字寻找合适的位置。

def find_all_missing_numbers(nums):

        i = 0 # Will happen only once so O(1)

        # no linear search for true position
        while i < len(nums):
            j = nums[i] - 1 # Will happen once per each iteration
            if nums[i] != nums[j]: # Condition Check Will happen once per iteration
                nums[i], nums[j] = nums[j], nums[i] # Will happen if if Condition is true so "c" times
            else: 
                i += 1 # Will happen if if Condition is not true so "c'" times

        missingNumbers = []

        # O(n)
        for i in range(len(nums)):
            if nums[i] != i + 1:
                missingNumbers.append(i + 1)

        return missingNumbers

所以:

1 + Len(nums)*(1 + 1 + c + c' + 1) + n

如果Len(nums)= n那么

1 + 3n + (c + c')n + n = 1 + (3+C)n + n ~ O(n)

暂无
暂无

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.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM