有没有办法规避Python list.append（）逐渐变慢在循环中的列表增长？ [英] Is there a way to circumvent Python list.append() becoming progressively slower in a loop as the list grows?

问题描述

我有一个大文件，我读取，并将每几行转换为一个对象的实例。

由于我循环的文件，我使用list.append（instance）将实例存储到一个列表中，然后继续循环。

这是一个约100MB的文件，因此它不会太大，但随着列表变大，循环逐渐减慢。 （我打印循环中每圈的时间）。

这不是循环的内在〜当我打印每个新的实例，当我循环通过文件，程序以恒定速度进行 - 它只是当我将它们附加到一个列表它变慢。

我的朋友建议在while循环之前禁用垃圾收集，然后启用它;进行垃圾回收调用。

有没有人观察到类似的问题，list.append越来越慢？有没有其他方法来绕过这个？

---

我会尝试以下两个建议。

（1）预分配内存〜最好的方法是什么？ （2）尝试使用deque

多个帖子（见Alex Martelli的评论）提示内存碎片化（他有大量的可用内存，像我一样）〜但没有明显修复此性能。

要复制这种现象，请运行下面提供的测试代码，并假设这些列表有有用的数据。

---

gc.disable（）和gc.enable（）有助于计时。

解决方案

您观察到的性能不佳是导致Python垃圾回收器中的错误。 要解决此问题，请在构建列表时停用垃圾回收，并在完成后将其打开。您会发现，性能接近于预期在Python中列出的附加的amoritized 0（1）行为。 em>

（你也可以调整垃圾收集器的触发器，或者随着进度选择性地调用collect，但我不会在这个答案中探索这些选项，因为它们更复杂，我怀疑您的使用案例符合上述解决方案。）

背景：

请参阅：

I have a big file I'm reading from, and convert every few lines to an instance of an Object.

Since I'm looping through the file, I stash the instance to a list using list.append(instance), and then continue looping.

This is a file that's around ~100MB so it isn't too large, but as the list grows larger, the looping slows down progressively. (I print the time for each lap in the loop).

This is not intrinsic to the loop ~ when I print every new instance as I loop through the file, the program progresses at constant speed ~ it is only when I append them to a list it gets slow.

My friend suggested disabling garbage collection before the while loop and enabling it afterward & making a garbage collection call.

Did anyone else observe a similar problem with list.append getting slower? Is there any other way to circumvent this?

---

I'll try the following two things suggested below.

(1) "pre-allocating" the memory ~ what's the best way to do this? (2) Try using deque

Multiple posts (see comment by Alex Martelli) suggested memory fragmentation (he has a large amount of available memory like I do) ~ but no obvious fixes to performance for this.

To replicate the phenomenon, please run the test code provided below in the answers and assume that the lists have useful data.

---

gc.disable() and gc.enable() helps with the timing. I'll also do a careful analysis of where all the time is spent.

解决方案

The poor performance you observe is caused by a bug in the Python garbage collector. To resolve this issue, disable garbage collection as you build the list and turn it on after you finish. You will find that performance approximates the amoritized 0(1) behavior expected of list appending in Python.

(You can also tweak the garbage collector's triggers or selectively call collect as you progress, but I do not explore these options in this answer because they are more complex and I suspect your use case is amenable to the above solution.)

Background:

See: http://bugs.python.org/issue4074 and also http://docs.python.org/release/2.5.2/lib/module-gc.html

The reporter observes that appending complex objects (objects that aren't numbers or strings) to a list slows linearly as the list grows in length.

The reason for this behavior is that the garbage collector is checking and rechecking every object in the list to see if they are eligible for garbage collection. This behavior causes the linear increase in time to add objects to a list. A fix is expected to land in py3k, so it should not apply to the interpreter you are using.

Test:

I ran a test to demonstrate this. For 1k iterations I append 10k objects to a list, and record the runtime for each iteration. The overall runtime difference is immediately obvious. With garbage collection disabled during the inner loop of the test, runtime on my system is 18.6s. With garbage collection enabled for the entire test, runtime is 899.4s.

This is the test:

import time
import gc

class A:
    def __init__(self):
        self.x = 1
        self.y = 2
        self.why = 'no reason'

def time_to_append(size, append_list, item_gen):
    t0 = time.time()
    for i in xrange(0, size):
        append_list.append(item_gen())
    return time.time() - t0

def test():
    x = []
    count = 10000
    for i in xrange(0,1000):
        print len(x), time_to_append(count, x, lambda: A())

def test_nogc():
    x = []
    count = 10000
    for i in xrange(0,1000):
        gc.disable()
        print len(x), time_to_append(count, x, lambda: A())
        gc.enable()

Full source: http://hypervolu.me/~erik/programming/python_lists/listtest.py.txt

Graphical result: Red is with gc on, blue is with gc off. y-axis is seconds scaled logarithmically.

As the two plots differ by several orders of magnitude in the y component, here they are independently with the y-axis scaled linearly.

Interestingly, with garbage collection off, we see only small spikes in runtime per 10k appends, which suggests that Python's list reallocation costs are relatively low. In any case, they are many orders of magnitude lower than the garbage collection costs.

The density of the above plots make it difficult to see that with the garbage collector on, most intervals actually have good performance; it's only when the garbage collector cycles that we encounter the pathological behavior. You can observe this in this histogram of 10k append time. Most of the datapoints fall around 0.02s per 10k appends.

The raw data used to produce these plots can be found at http://hypervolu.me/~erik/programming/python_lists/

这篇关于有没有办法规避Python list.append（）逐渐变慢在循环中的列表增长？的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！