英文:
Convert ndarray to 2D ndarray
问题
给定一个ndarray：
import numpy as npimport randomidxs = np.arange(10)np.random.shuffle(idxs)
我得到一个输出：
array([1, 2, 7, 8, 5, 9, 3, 4, 6, 0])
我想将其转换为一个ndarrays数组，看起来像这样：
array([[1, 2, 3, 4, 5],[2, 3, 4, 5, 6],[7, 8, 9, 10, 11],[8, 9, 10, 11, 12],...])
对于第一个数组的每个元素，我希望获取其上方的5个索引。我应该如何做？我尝试使用以下代码：
np.reshape(idxs, (idxs.size, 1))
将其转换为以下形式：
array([[1],[2],[7],[8],...])
但之后我想获取所有上方的5个索引，如上所述。感谢任何帮助。
英文:
Given an ndarray:
import numpy as npimport randomidxs = np.arange(10)np.random.shuffle(idxs)
I get an output:
array([1, 2, 7, 8, 5, 9, 3, 4, 6, 0])
I would like to convert this to an array of ndarrays that looks like:
array([[1, 2, 3, 4, 5],[2, 3, 4, 5, 6],[7, 8, 9, 10, 11],[8, 9, 10, 11, 12],...])
where for each element of the first array, I want the indexes up to 5 above it. How can I do this? I have tried using
np.reshape(idxs, (idxs.size, 1))
to get it into this:
array([[1],[2],[7],[8],...])
but then I would like to get all indexes up to 5 above it, as mentioned above.
I hope I explained this clearly, grateful for any help.
答案1
得分: 2
使用广播：
idxs = np.array([1, 2, 7, 8, 5, 9, 3, 4, 6, 0])out = idxs[:, None] + np.arange(5)# 或者out = idxs.reshape((-1, 1)) + np.arange(5)
或者（主要是为了好玩，因为这不够高效），如果您的索引是0到n：
from numpy.lib.stride_tricks import sliding_window_view as swvout = swv(np.arange(10 + 5), 5)[idxs]
（如果不是，这将不必要地生成许多中间行）
输出：
array([[ 1, 2, 3, 4, 5],[ 2, 3, 4, 5, 6],[ 7, 8, 9, 10, 11],[ 8, 9, 10, 11, 12],[ 5, 6, 7, 8, 9],[ 9, 10, 11, 12, 13],[ 3, 4, 5, 6, 7],[ 4, 5, 6, 7, 8],[ 6, 7, 8, 9, 10],[ 0, 1, 2, 3, 4]])
英文:
Use broadcasting:
idxs = np.array([1, 2, 7, 8, 5, 9, 3, 4, 6, 0])out = idxs[:,None] + np.arange(5)# orout = idxs.reshape((-1, 1))+np.arange(5)
Alternatively (mostly for fun as this is not as efficient), if your indices are 0-n:
from numpy.lib.stride_tricks import sliding_window_view as swvout = swv(np.arange(10+5), 5)[idxs]
(If not, this would generate many intermediate rows unnecessarily)
Output:
array([[ 1, 2, 3, 4, 5],[ 2, 3, 4, 5, 6],[ 7, 8, 9, 10, 11],[ 8, 9, 10, 11, 12],[ 5, 6, 7, 8, 9],[ 9, 10, 11, 12, 13],[ 3, 4, 5, 6, 7],[ 4, 5, 6, 7, 8],[ 6, 7, 8, 9, 10],[ 0, 1, 2, 3, 4]])
答案2
得分: 0
我可以帮您翻译如下内容：
"I have this simple way of getting a (n, 1) shaped nd-array out of your code by using the np.newaxis."
"我有一种简单的方法，可以通过使用np.newaxis从您的代码中获取一个(n, 1)形状的nd-array。"
英文:
I have this simple way of getting a (n, 1) shaped nd-array out of your code by using the np.newaxis.
In [3]: np.arange(10)[:, np.newaxis]Out[3]:array([[0],[1],[2],[3],[4],[5],[6],[7],[8],[9]])
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论