我已经导入了PyTorch所需的所有库。这是从我的图像获得的numpy数组。我的图片尺寸是150x150灰度图片
array([[array([[ 89, 117, 59, ..., 39, 48, 38],
[118, 85, 96, ..., 41, 42, 31],
[171, 118, 70, ..., 33, 42, 32],
...,
[ 32, 49, 32, ..., 153, 181, 180],
[ 30, 25, 28, ..., 150, 180, 185],
[ 48, 33, 34, ..., 159, 168, 184]], dtype=uint8),
array([0., 1.])],
[array([[195, 192, 202, ..., 255, 254, 255],
[206, 190, 200, ..., 255, 255, 255],
[221, 186, 192, ..., 254, 254, 255],
...,
[254, 255, 255, ..., 254, 253, 255],
[255, 255, 255, ..., 255, 255, 255],
[255, 255, 255, ..., 255, 255, 255]], dtype=uint8),
array([0., 1.])],
[array([[60, 58, 66, ..., 10, 11, 10],
[61, 59, 56, ..., 11, 11, 11],
[58, 53, 63, ..., 10, 11, 10],
...,
[23, 45, 56, ..., 46, 38, 19],
[71, 46, 91, ..., 34, 40, 36],
[69, 76, 32, ..., 65, 26, 51]], dtype=uint8),
array([1., 0.])],
...,
[array([[247, 234, 231, ..., 112, 112, 114],
[243, 229, 230, ..., 113, 114, 114],
[236, 228, 231, ..., 118, 117, 115],
...,
[ 22, 24, 24, ..., 57, 61, 61],
[ 28, 26, 25, ..., 88, 87, 86],
[ 30, 28, 29, ..., 92, 91, 91]], dtype=uint8),
array([0., 1.])],
[array([[207, 194, 195, ..., 254, 254, 255],
[209, 199, 200, ..., 254, 255, 255],
[199, 207, 201, ..., 254, 254, 255],
...,
[255, 255, 255, ..., 255, 255, 255],
[255, 255, 255, ..., 255, 255, 255],
[255, 255, 255, ..., 255, 255, 255]], dtype=uint8),
array([1., 0.])],
[array([[ 65, 62, 72, ..., 54, 69, 53],
[ 80, 44, 35, ..., 60, 37, 37],
[ 75, 45, 93, ..., 43, 52, 54],
...,
[ 39, 41, 106, ..., 21, 19, 18],
[ 49, 34, 106, ..., 19, 18, 20],
[ 42, 45, 88, ..., 19, 15, 17]], dtype=uint8),
array([1., 0.])]], dtype=object)
数据类型为对象，我收到此错误：
TypeError：无法转换numpy.object_类型的np.ndarray。唯一受支持的类型为：float64，float32，float16，int64，int32，int16，int8，uint8和bool。
如何将该对象dtype转换为受支持的类型？