您正在寻找以下内容：

import numpy as np

u1 = np.array([[[-2.04853845, -1.47283101,  1.],
                [-2.05009646, -1.46780913,  1.],
                [-2.05165539, -1.46278426,  1.]],
               [[-4.94165412,  7.8524744,   1.],
                [ 0.94540647,  0.86456925,  1.],
                [-4.94916228,  7.87667525,  1.]]])

np.logical_and(u1[:,:,0:2]>0, u1[:,:,0:2]<1)
array([[[False, False],
        [False, False],
        [False, False]],

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

字符串

提取元素并与您的值进行比较：

import numpy as np

u1 = np.array([[[-2.04853845, -1.47283101,  1.        ],
                [-2.05009646, -1.46780913,  1.        ],
                [-2.05165539, -1.46278426,  1.        ]],
               [[-4.94165412,  7.8524744 ,  1.        ],
                [ 0.94540647,   0.86456925,  1.       ],
                [-4.94916228,  7.87667525,  1.        ]]])

first = (u1[:, :, 0] > 0) * (u1[:, :, 0] < 1)
second = (u1[:, :, 1] > 0) * (u1[:, :, 1] < 1)

condition = first * second
# Condition is:
# [[False False False]
# [False  True False]]

字符串
然后提取您感兴趣的行：

values = u1[condition]

print(values)
# Prints:
# [[0.94540647 0.86456925 1.        ]]

型

定义以下函数，检查行的前两个元素：

def myfunc(row):
    return np.logical_and(row[0:2] >= 0, row[0:2] <= 1).all()

字符串
然后将其应用于最后一个维度沿着的每个一维切片（我们称之为 * 行 *）：

np.apply_along_axis( myfunc, axis=-1, arr=u1 )

型
示例数据的结果为：

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

型