堆排序解决 top k 问题

最新推荐文章于 2024-09-05 14:36:31 发布

bdss58

最新推荐文章于 2024-09-05 14:36:31 发布

阅读量1k

点赞数

CC 4.0 BY-SA版权

分类专栏：算法文章标签：堆排序

本文链接：https://blog.youkuaiyun.com/bdss58/article/details/54233440

算法专栏收录该内容

23 篇文章

订阅专栏

本文介绍了一种使用堆排序改进线性搜索效率的方法，通过对比线性搜索与堆排序搜索找到数组中最大的N个数的过程及耗时，展示了堆排序搜索在大数据集上的优势。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

import time
import heapq
import random

def gen_large_arr(n):
    arr = [i for i in range(n)]
    random.shuffle(arr)
    return arr

# find top n item from bigarray using liner search
def liner_search(bigarray, n):
    return sorted(bigarray, reverse=True)[:n]

# using heap sort
def heap_search(bigarray, n):
    heap = []
    for elem in bigarray:
        if len(heap) < n or elem > heap[0]:
            if len(heap) == n:
                heapq.heappop(heap)
            heapq.heappush(heap, elem)

    return heap


start = time.time()
bigarray = gen_large_arr(10 * 1000 * 1000)
print 'gen arr took %g s' % (time.time() - start)

start = time.time()
print liner_search(bigarray, 10)
print 'linear search took %g s' % (time.time() - start)

start = time.time()
print heap_search(bigarray, 10)
print 'heap search took %g s' %(time.time() - start)