纠正错误：用作索引的数组必须是整数（或布尔）类型

Question

这是我的代码：

np.random.seed(0)

luck_A = np.random.uniform(0, 1, 大小 = (2, 1, 100))

A = np.random.uniform(0, 1, 大小 = (2, 1, 100))

index_select_A = (np.argpartition(A, -10)[:, :, -10:])

结果 = np.apply_along_axis(lambda x: lucky_A[x, :, index_select_A[x]], axis=2, arr=luck_A)

我想创建一个数组，其中包含luck_A的元素，但仅包含索引在index_select_A中的元素。但是当我运行代码时，它不断给出错误：用作索引的数组必须是整数（或布尔）类型。但index_select_A中的元素是整数

我已经尝试过 .astype(int)。没成功

Answer 1

您的错误：

Cell In[3], line 9, in <lambda>(x)
      5 A = np.random.uniform(0, 1, size = (2, 1, 100))
      7 index_select_A = (np.argpartition(A, -10)[:, :, -10:])
----> 9 result = np.apply_along_axis(lambda x: luck_A[x, :, index_select_A[x]], axis=2, arr=luck_A)

IndexError: arrays used as indices must be of integer (or boolean) type

index_select_A[x]

突出显示。

重写 lambda 以允许进行一些打印：

In [4]: def foo(x):
   ...:     print(x)
   ...:     print(index_select_A[x])
   ...:     return luck_A[x, :, index_select_A[x]]
   ...:     

In [5]: np.apply_along_axis(foo, axis=2, arr =luck_A)
[0.5488135  0.71518937 0.60276338 0.54488318 0.4236548  0.64589411
 0.43758721 0.891773   0.96366276 0.38344152 0.79172504 0.52889492
 0.56804456 0.92559664 0.07103606 0.0871293  0.0202184  0.83261985

...] -------------------------------------------------- ------------------------ IndexError Traceback（最近一次调用最后一次） 第 1 行 [5] 中的单元格 ----> 1 np.apply_along_axis(foo, axis=2, arr =luck_A)

File <__array_function__ internals>:200, in apply_along_axis(*args, **kwargs)

File ~\miniconda3\lib\site-packages\numpy\lib\shape_base.py:379, in apply_along_axis(func1d, axis, arr, *args, **kwargs)
    375 except StopIteration as e:
    376     raise ValueError(
    377         'Cannot apply_along_axis when any iteration dimensions are 0'
    378     ) from None
--> 379 res = asanyarray(func1d(inarr_view[ind0], *args, **kwargs))
    381 # build a buffer for storing evaluations of func1d.
    382 # remove the requested axis, and add the new ones on the end.
    383 # laid out so that each write is contiguous.
    384 # for a tuple index inds, buff[inds] = func1d(inarr_view[inds])
    385 buff = zeros(inarr_view.shape[:-1] + res.shape, res.dtype)

Cell In[4], line 3, in foo(x)
      1 def foo(x):
      2     print(x)
----> 3     print(index_select_A[x])
      4     return luck_A[x, :, index_select_A[x]]

IndexError: arrays used as indices must be of integer (or boolean) type

apply

将

luck_A

的一行形状 (100,) 传递给函数（如

x

）。这是一个浮点数数组。当然不能作为索引！

In [6]: luck_A.shape
Out[6]: (2, 1, 100)