PyTorch基础——均值、（无偏估计）方差

import torch 

dat = torch.Tensor([1,2,3])

# 1.求均值
dat_mean = torch.mean(dat) # => dat_mean = tensor(2.)

# 2.求方差
dat_var1 = torch.var(dat)                  # => dat_var1 = tensor(1.),对应于无偏估计
dat_var2 = torch.var(dat, unbiased=False)  # => dat_var2 = tensor(0.6667)

torch.var(....)函数中unbiased默认为True，表示方差的无偏估计，除数为 —— 元素个数减一

dat_var1（无偏估计）计算：
( (1-2)*(1-2) +(2-2)*(2-2)+(3-2)*(3-2) ) / (3-1) = 2/2 = 1

dat_var2计算：
( (1-2)*(1-2)+(2-2)*(2-2)+(3-2)*(3-2) )  / 3 = 2/3 = 0.6666667