如何匹配两个numpy数组中包含的值对 [英] How to match pairs of values contained in two numpy arrays

问题描述

我有两组坐标，想找出 coo 组的哪些坐标与 targets 组中的任何坐标相同.我想知道 coo 集中的索引，这意味着我想获取一个索引列表或布尔值.

I have two sets of coordinates and want to find out which coordinates of the coo set are identical to any coordinate in the targets set. I want to know the indices in the coo set which means I'd like to get a list of indices or of bools.

import numpy as np

coo = np.array([[1,2],[1,6],[5,3],[3,6]]) # coordinates
targets = np.array([[5,3],[1,6]]) # coordinates of targets

print(np.isin(coo,targets))

[[ True False]
 [ True  True]
 [ True  True]
 [ True  True]]

期望的结果将是以下两个之一:

The desired result would be one of the following two:

[False True True False] # bool list
[1,2] # list of concerning indices

我的问题是...

• np.isin 没有 axis 属性，因此我可以使用 axis = 1 .
• 即使对输出的每一行应用 logical和，对于最后一个元素也会返回 True ，这是错误的.

• np.isin has no axis-attribute so that I could use axis=1.
• even applying logical and to each row of the output would return True for the last element, which is wrong.

我知道循环和条件，但是我确信Python具备了更优雅的解决方案的方法.

I am aware of loops and conditions but I am sure Python is equipped with ways for a more elegant solution.

推荐答案

对于大数组，此解决方案的伸缩性将变差，在这种情况下，其他建议的答案将表现更好.

This solution will scale worse for large arrays, for such cases the other proposed answers will perform better.

这是利用 广播的一种方法> :

Here's one way taking advantage of broadcasting:

(coo[:,None] == targets).all(2).any(1)
# array([False,  True,  True, False])

---

详细信息

通过直接比较，将第一个轴添加到 coo 中，检查 coo 中的每一行是否与 target 中的另一行相匹配可以针对目标进行广播:

Check for every row in coo whether or not it matches another in target by direct comparisson having added a first axis to coo so it becomes broadcastable against targets:

(coo[:,None] == targets)

array([[[False, False],
        [ True, False]],

       [[False, False],
        [ True,  True]],

       [[ True,  True],
        [False, False]],

       [[False, False],
        [False,  True]]])

然后检查第二个轴上哪个 ndarrays 具有 所有 值转换为 True :

Then check which ndarrays along the second axis have all values to True:

(coo[:,None] == targets).all(2)

array([[False, False],
       [False,  True],
       [ True, False],
       [False, False]])

最后使用 any 检查哪些行至少具有一个 True .

这篇关于如何匹配两个numpy数组中包含的值对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！