带有异常的Python随机样本 [英] Python random sample with exception

问题描述

我正在尝试创建myList的示例，但i_除外.

I'm attempting to create a sample of myList with the exception of i_.

我的尝试:

i_ = 'd'

myList = ['a','b','c','d','e']

myList = random.sample(myList, 3)

尽管上面的示例工作正常，但仍有可能从myList中删除i_.

Although the above sample works fine, there is still a chance that i_ is removed from myList.

我想创建一个样本，而不会删除i_的值.

I would like to create a sample without the value of i_ being removed.

所需的输出:

Desired Output:

任何包含3个项目的列表输出，包括i_，例如:

Any list output containing 3 items including i_, e.g:

['a','d','e']

推荐答案

仅从列表中抽取2个值，而不使用i_并在以后插入i_:

Just sample 2 values from the list without the i_ and insert i_ later:

new_list = random.sample([i for i in myList if i != i_], 2)
new_list.insert(random.randrange(0, 3), i_)

但是，这假设i_在您的列表中仅出现一次-在您的示例中似乎是合理的，但出于完整性考虑，我想提及这一点.另外，如果列表中有多个i_，我也不确定期望的结果是什么.

But this assumes that i_ occurs only once in your list - seems reasonable given your example but I wanted to mention this for completeness. Also I'm not sure what the desired result should be if there were multiple i_ in the list.

您还可以使用hit& miss方法来生成示例，直到获得包含i_的示例:

You could also use a hit&miss approach that generates a sample until you get one that contains i_:

new_list = []
while i_ not in new_list:
    new_list = random.sample(myList, 3)

请注意，如果不太可能在示例中绘​​制i_，则这可能会非常慢.对于5个元素的3个样本，它相当快，但是如果从1000个元素中绘制3个样本，则它可能会真的很慢.

Note that this could be really slow if it's unlikely that i_ is drawn in the sample. For a 3-sample of 5 elements it's quite fast but it could be really slow if you draw a 3-sample from 1000 elements.

这篇关于带有异常的Python随机样本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！