http://www.coder4.com/archives/3844
需1求:给出N长的序列,求出TopK大的元素,使用小顶堆,heapq模块实现。
import heapq
import random
class TopkHeap(object):
def __init__(self, k):
self.k = k
self.data = []
def Push(self, elem):
if len(self.data) < self.k:
heapq.heappush(self.data, elem)
else:
topk_small = self.data[0]
if elem > topk_small:
heapq.heapreplace(self.data, elem)
def TopK(self):
return [x for x in reversed([heapq.heappop(self.data) for x in xrange(len(self.data))])]
if __name__ == "__main__":
print "Hello"
list_rand = random.sample(xrange(1000000), 100)
th = TopkHeap(3)
for i in list_rand:
th.Push(i)
print th.TopK()
print sorted(list_rand, reverse=True)[0:3]
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 |
import heapq import random class TopkHeap(object): def __init__(self, k): self.k = k self.data = [] def Push(self, elem): if len(self.data) < self.k: heapq.heappush(self.data, elem) else: topk_small = self.data[0] if elem > topk_small: heapq.heapreplace(self.data, elem) def TopK(self): return [x for x in reversed([heapq.heappop(self.data) for x in xrange(len(self.data))])] if __name__ == "__main__": print "Hello" list_rand = random.sample(xrange(1000000), 100) th = TopkHeap(3) for i in list_rand: th.Push(i) print th.TopK() print sorted(list_rand, reverse=True)[0:3] |
上面的用heapq就能轻松搞定。
变态的需求来了:给出N长的序列,求出BtmK小的元素,即使用大顶堆。
heapq在实现的时候,没有给出一个类似Java的Compartor函数接口或比较函数,开发者给出了原因见这里:http://code.activestate.com/lists/python-list/162387/
于是,人们想出了一些很NB的思路,见:http://stackoverflow.com/questions/14189540/python-topn-max-heap-use-heapq-or-self-implement
我来概括一种最简单的:
将push(e)改为push(-e)、pop(e)改为-pop(e)。
也就是说,在存入堆、从堆中取出的时候,都用相反数,而其他逻辑与TopK完全相同,看代码:
class BtmkHeap(object):
def __init__(self, k):
self.k = k
self.data = []
def Push(self, elem):
# Reverse elem to convert to max-heap
elem = -elem
# Using heap algorighem
if len(self.data) < self.k:
heapq.heappush(self.data, elem)
else:
topk_small = self.data[0]
if elem > topk_small:
heapq.heapreplace(self.data, elem)
def BtmK(self):
return sorted([-x for x in self.data])
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 |
class BtmkHeap(object): def __init__(self, k): self.k = k self.data = [] def Push(self, elem): # Reverse elem to convert to max-heap elem = -elem # Using heap algorighem if len(self.data) < self.k: heapq.heappush(self.data, elem) else: topk_small = self.data[0] if elem > topk_small: heapq.heapreplace(self.data, elem) def BtmK(self): return sorted([-x for x in self.data]) |
经过测试,是完全没有问题的,这思路太Trick了……