如何创建任意长度字符串的 numpy 数组? [英] How to create a numpy array of arbitrary length strings?

问题描述

我是 Python 的完全新手，但似乎给定的字符串能够(有效地)是任意长度.即，您可以使用 string str 并不断添加:str += "some stuff...".有没有办法制作这样的字符串数组?

当我尝试这个时，每个元素只存储一个字符

strArr = numpy.empty(10, dtype='string')对于我在范围内(0,10)strArr[i] = "测试"

另一方面，我知道我可以初始化一个特定长度字符串的数组，即

strArr = numpy.empty(10, dtype='s256')

最多可存储 10 个字符串，最多 256 个字符.

解决方案

您可以通过创建 dtype=object 数组来实现.如果您尝试将一个长字符串分配给普通的 numpy 数组，它会截断该字符串:

<预><代码>>>>a = numpy.array(['apples', 'foobar', 'cowboy'])>>>a[2] = '香蕉'>>>一种数组(['苹果'，'foobar'，'香蕉']，dtype='|S6')

但是当您使用 dtype=object 时，您会得到一个 Python 对象引用数组.所以你可以拥有python字符串的所有行为:

<预><代码>>>>a = numpy.array(['apples', 'foobar', 'cowboy'], dtype=object)>>>一种数组([苹果，foobar，牛仔]，dtype=object)>>>a[2] = '香蕉'>>>一种数组([苹果，foobar，香蕉]，dtype=object)

确实，因为它是一个对象数组，您可以将任何种类的python对象分配给该数组:

<预><代码>>>>a[2] = {1:2, 3:4}>>>一种数组([苹果，foobar，{1:2，3:4}]，dtype=object)

然而，这抵消了使用 numpy 的许多好处，它如此之快，因为它适用于大量连续的原始内存块.使用 python 对象会增加很多开销.一个简单的例子:

<预><代码>>>>a = numpy.array(['abba' for _ in range(10000)])>>>b = numpy.array(['abba' for _ in range(10000)], dtype=object)>>>%timeit a.copy()100000 个循环，最好的 3 个:每个循环 2.51 us>>>%timeit b.copy()10000 个循环，最好的 3 个:每个循环 48.4 us

I'm a complete rookie to Python, but it seems like a given string is able to be (effectively) arbitrary length. i.e. you can take a string str and keeping adding to it: str += "some stuff...". Is there a way to make an array of such strings?

When I try this, each element only stores a single character

strArr = numpy.empty(10, dtype='string')
for i in range(0,10)
    strArr[i] = "test"

On the other hand, I know I can initialize an array of certain length strings, i.e.

strArr = numpy.empty(10, dtype='s256')

which can store 10 strings of up to 256 characters.

解决方案

You can do so by creating an array of dtype=object. If you try to assign a long string to a normal numpy array, it truncates the string:

>>> a = numpy.array(['apples', 'foobar', 'cowboy'])
>>> a[2] = 'bananas'
>>> a
array(['apples', 'foobar', 'banana'], 
      dtype='|S6')

But when you use dtype=object, you get an array of python object references. So you can have all the behaviors of python strings:

>>> a = numpy.array(['apples', 'foobar', 'cowboy'], dtype=object)
>>> a
array([apples, foobar, cowboy], dtype=object)
>>> a[2] = 'bananas'
>>> a
array([apples, foobar, bananas], dtype=object)

Indeed, because it's an array of objects, you can assign any kind of python object to the array:

>>> a[2] = {1:2, 3:4}
>>> a
array([apples, foobar, {1: 2, 3: 4}], dtype=object)

However, this undoes a lot of the benefits of using numpy, which is so fast because it works on large contiguous blocks of raw memory. Working with python objects adds a lot of overhead. A simple example:

>>> a = numpy.array(['abba' for _ in range(10000)])
>>> b = numpy.array(['abba' for _ in range(10000)], dtype=object)
>>> %timeit a.copy()
100000 loops, best of 3: 2.51 us per loop
>>> %timeit b.copy()
10000 loops, best of 3: 48.4 us per loop

这篇关于如何创建任意长度字符串的 numpy 数组?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！