numpy：找到给定索引数组的行的值 [英] Numpy: Find the values of rows given an array of indexes

问题描述

我有一个2D值数组和一个1D索引数组。我想使用索引数组从每一行的索引中提取值。以下代码将成功完成此操作：

I have a 2D array of values and a 1D array of indexes. I want to pull the values from the index of each row using an array of indexes. The following code would do this successfully:

from pprint import pprint
import numpy as np
_2Darray = np.arange(100, dtype = np.float16)
_2Darray = _2Darray.reshape((10, 10))
array_indexes = [5,5,5,4,4,4,6,6,6,8]
index_values = []
for row, index in enumerate(array_indexes):
    index_values.append(_2Darray[row, index])
pprint(_2Darray)
print index_values

返回

array([[  0.,   1.,   2.,   3.,   4.,   5.,   6.,   7.,   8.,   9.],
       [ 10.,  11.,  12.,  13.,  14.,  15.,  16.,  17.,  18.,  19.],
       [ 20.,  21.,  22.,  23.,  24.,  25.,  26.,  27.,  28.,  29.],
       [ 30.,  31.,  32.,  33.,  34.,  35.,  36.,  37.,  38.,  39.],
       [ 40.,  41.,  42.,  43.,  44.,  45.,  46.,  47.,  48.,  49.],
       [ 50.,  51.,  52.,  53.,  54.,  55.,  56.,  57.,  58.,  59.],
       [ 60.,  61.,  62.,  63.,  64.,  65.,  66.,  67.,  68.,  69.],
       [ 70.,  71.,  72.,  73.,  74.,  75.,  76.,  77.,  78.,  79.],
       [ 80.,  81.,  82.,  83.,  84.,  85.,  86.,  87.,  88.,  89.],
       [ 90.,  91.,  92.,  93.,  94.,  95.,  96.,  97.,  98.,  99.]], dtype=float16)
[5.0, 15.0, 25.0, 34.0, 44.0, 54.0, 66.0, 76.0, 86.0, 98.0]

但是我只想使用numpy函数。我已经尝试了很多numpy函数，但是似乎没有一个函数可以简单地完成这项任务。

But I want to do it using only numpy functions. I have tried a whole bunch of numpy functions, but none of them seem to do this fairly simply task.

预先感谢！

编辑
我设法弄清楚我的实现是：
V_high = np.fromiter（ （

Edit I managed to figure out what my implementation would be: V_high = np.fromiter((

index_values = _2Darray[ind[0], ind[1]] for ind in
                    enumerate(array_indexes)),
                    dtype = _2Darray.dtype,
                    count = len(_2Darray))

感谢扎根，我已经完成了两个实现。现在进行一些分析：
我的实现通过cProfiler

Thanks to root I've got both implementations worked out. Now for some profiling: My implementation run through cProfiler

   ncalls  tottime  percall  cumtime  percall filename:lineno(function)
    2    0.274    0.137    0.622    0.311 {numpy.core.multiarray.fromiter}
20274    0.259    0.000    0.259    0.000 lazer_np.py:86(<genexpr>)

和root：

    4    0.000    0.000    0.000    0.000 {numpy.core.multiarray.array}
    1    0.000    0.000    0.000    0.000 {numpy.core.multiarray.arange}

我不敢相信，但是cProfiler根本没有检测到root需要花费任何时间的方法。我认为这一定是某种错误，但是肯定明显更快。在较早的测试中，我得到根的速度大约快3倍

注意：这些测试是在np的shape =（20273，200）数组上完成的.float16值。另外，每个索引必须为每个测试运行两次。

Note: these tests were done on a shape = (20273, 200) array of np.float16 values. Additionally, each indexing had to be run twice for each test.

推荐答案

这应该做到：

row = numpy.arange(_2Darray.shape[0])
index_values = _2Darray[row, array_indexes]

Numpy允许您使用两个数组对2d数组（或nd数组）进行索引，例如：

Numpy allows you to index 2d arrays (or nd arrays really) with two arrays such that:

for i in range(len(row)):
    result1[i] = array[row[i], col[i]]

result2 = array[row, col]
numpy.all(result1 == result2)

这篇关于numpy：找到给定索引数组的行的值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！