给你一个用字符数组 tasks
表示的 CPU 需要执行的任务列表,用字母 A 到 Z 表示,以及一个冷却时间 n
。每个周期或时间间隔允许完成一项任务。任务可以按任何顺序完成,但有一个限制:两个 相同种类 的任务之间必须有长度为 n
的冷却时间。
返回完成所有任务所需要的 最短时间间隔 。
示例 1:
输入:tasks = ["A","A","A","B","B","B"], n = 2 输出:8 解释:A -> B -> (待命) -> A -> B -> (待命) -> A -> B 在本示例中,两个相同类型任务之间必须间隔长度为 n = 2 的冷却时间,而执行一个任务只需要一个单位时间,所以中间出现了(待命)状态。
示例 2:
输入:tasks = ["A","A","A","B","B","B"], n = 0 输出:6 解释:在这种情况下,任何大小为 6 的排列都可以满足要求,因为 n = 0 ["A","A","A","B","B","B"] ["A","B","A","B","A","B"] ["B","B","B","A","A","A"] ... 诸如此类
示例 3:
输入:tasks = ["A","A","A","A","A","A","B","C","D","E","F","G"], n = 2 输出:16 解释:一种可能的解决方案是: A -> B -> C -> A -> D -> E -> A -> F -> G -> A -> (待命) -> (待命) -> A -> (待命) -> (待命) -> A
提示:
1 <= tasks.length <= 104
tasks[i]
是大写英文字母0 <= n <= 100
思路:
首先分析题意,由于每种工作有CD,所以当我们可以做很多种工作的时候,我们应该优先选择剩余更多的工作,这里就提示需要使用最大堆来动态找最大值。
其次,当工作在CD的时候,我们是需要单独考虑它的,只有CD好了才能继续干活。由于时间是线性的,这里可以用队列来存放正在CD中的工作。
总结:
1. 使用最大堆存放当前不在CD里的工作,key是工作的频率
2. 使用队列存放正在CD的工作,队列的每个元素是(freq, ready_time)
3. 正确地在这两个数据结构间转移工作
时间复杂度:O(NlogN)
空间复杂度:O(N)
from heapq import *
class Solution:
def leastInterval(self, tasks: List[str], n: int) -> int:
# we should prioritize working on tasks of higher frequency by greedy
# use a max_heap to track all workable(not in cooldown) tasks since we need to sort by frequency
# use a queue to track all in colldown tasks
# periodically move tasks between these two data structures
time = 0
max_heap = [-freq for freq in Counter(tasks).values()]
heapify(max_heap)
queue = deque() # (freq, ready_time)
while max_heap or queue:
time += 1
if not max_heap:
# no tasks avaiable now, we must wait till the next job become ready
time = queue[0][1]
else:
# do tasks in the max_heap
freq = -heappop(max_heap)
if freq > 1:
queue.append((freq - 1, time + n))
# check if we can move the first task in queue to the max_heap
if queue and queue[0][1] <= time:
heappush(max_heap, -queue.popleft()[0])
return time