函数用法

官方文档给的用法：

(means, std, out=None)

意思为means给定tensor的均值范围和形状，std给出每个均值服从的标准差

官方示例：

(means=(1, 11), std=(1, 0, -0.1))

 1.5104
 1.6955
 2.4895
 4.9185
 4.9895
 6.9155
 7.3683
 8.1836
 8.7164
 9.8916
[ of size 10]

实际情况：

>>> (means=(1, 11), std=(1, 0, -0.1))
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: normal() received an invalid combination of arguments - got (means=Tensor, std=Tensor, ), but expected one of:
 * (Tensor mean, Tensor std, *,  generator, Tensor out)
 * (Tensor mean, float std, *,  generator, Tensor out)
 * (float mean, Tensor std, *,  generator, Tensor out)
 * (float mean, float std, tuple of ints size, *,  generator, Tensor out,  dtype,  layout,  device, bool pin_memory, bool requires_grad)

从错误类型中看出现在torch中已经不再使用means而是使用mean，即只能生成固定均值的正态分布。

正确的打开方式：

采用第四种：

(mean, std, size)

三个参数分别为均值，标准差和size

>>> (3, 0.1, (3, 4))
tensor([[2.9425, 3.1877, 2.9735, 3.0982],
        [3.0061, 2.9918, 2.7953, 3.0066],
        [2.8219, 2.9578, 2.8813, 2.9014]])

采用第三种：

(mean, stds)

两个参数分别为:均值和标准差，使用标准差来确定范围size

>>> (3, (3, 4)/10)
tensor([[2.8491, 3.0263, 3.0888, 3.0818],
        [3.1101, 2.7490, 3.1847, 3.0861],
        [2.8530, 2.8666, 2.9634, 3.1875]])