Numpy ndarray及Pytorch tensor中元素类型转换

ndarray需要使用对数组对象进行astype()操作进行类型转换

tensor则是使用type()

演示如下

import numpy as np
import torch

data1 = torch.randint(low = 0,high=10,size=(5,))
data2 = np.random.randint(0,10,5)
print(data1,end='\n\n')
print(data2,end='\n\n')
print(data1.type(torch.float32))
print(data2.astype(np.float32))

# tensor([6, 1, 2, 3, 9])
# 
# [4 9 4 7 9]
# 
# tensor([6., 1., 2., 3., 9.])
# [4. 9. 4. 7. 9.]