原题

在一个数组中找到前K大的数

样例
给出 [3,10,1000,-99,4,100], k = 3.
返回 [1000, 100, 10]

解题思路

• 方法一：快速排序，然后取出前k大的数。时间复杂度O(n*logn + k)

• 方法二：维护一个大小为k的最大堆/最小堆，代码如下。时间复杂度为O(n * logk)

• 关于Heap

• 在python中有两个接口：heapq和Queue.PriorityQueue。其中PriorityQueue module is using the heapq module which is slower because it adds locks, encapsulation, and a nice object oriented API.

• heapq的使用：

  • heapq.heappush: Push the value item onto the heap, maintaining the heap invariant.
  • heapq.heappop: Pop and return the smallest item from the heap, maintaining the heap invariant.
  • heapq.heapify: Transform list x into a heap, in-place, in linear time.

• Queue.PriorityQueue的使用 (myqueue = Queue.PriorityQueue)

  • myqueue.put()
  • myqueue.get()
  • myqueue.qsize()
  • myqueue.empty()

完整代码

import Queue

class Solution:
    '''
    @param {int[]} nums an integer array
    @param {int} k an integer
    @return {int[]} the top k largest numbers in array
    '''
    def topk(self, nums, k):
        # Write your code here
        res = []
        if not nums or k == 0:
            return res
            
        MaxHeap = Queue.PriorityQueue()
        for num in nums:
            MaxHeap.put(-num)
        
        for i in range(k):
            res.append(-MaxHeap.get())
        
        return res