从numpy数组中删除选择索引处的行 [英] Delete rows at select indexes from a numpy array
问题描述
在我的数据集中,我已经接近200行,但是为了最小的工作,例如,让我们假设以下数组:
In my dataset I've close to 200 rows but for a minimal working e.g., let's assume the following array:
arr = np.array([[1,2,3,4], [5,6,7,8],
[9,10,11,12], [13,14,15,16],
[17,18,19,20], [21,22,23,24]])
我可以对以下3行进行随机采样:
I can take a random sampling of 3 of the rows as follows:
indexes = np.random.choice(np.arange(arr.shape[0]), int(arr.shape[0]/2), replace=False)
使用这些索引,我可以如下选择测试用例:
Using these indexes, I can select my test cases as follows:
testing = arr[indexes]
我想删除这些索引处的行,并且可以将其余元素用于我的训练集.
I want to delete the rows at these indexes and I can use the remaining elements for my training set.
从帖子此处,看来training = np.delete(arr, indexes)
应该这样做.但是我却得到一维数组.
From the post here, it seems that training = np.delete(arr, indexes)
ought to do it. But I get 1d array instead.
我也在此处使用training = arr[indexes.astype(np.bool)]
,但没有给出清晰的分隔.我在训练和测试集中都得到了元素[5,6,7,8].
I also tried the suggestion here using training = arr[indexes.astype(np.bool)]
but it did not give a clean separation. I get element [5,6,7,8] in both the training and testing sets.
training = arr[indexes.astype(np.bool)]
testing
Out[101]:
array([[13, 14, 15, 16],
[ 5, 6, 7, 8],
[17, 18, 19, 20]])
training
Out[102]:
array([[ 1, 2, 3, 4],
[ 5, 6, 7, 8],
[ 9, 10, 11, 12]])
知道我在做什么错吗?谢谢.
Any idea what I am doing wrong? Thanks.
推荐答案
要从numpy数组中删除索引行:
To delete indexed rows from numpy array:
arr = np.delete(arr, indexes, axis=0)
这篇关于从numpy数组中删除选择索引处的行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!