Learn practical skills, build real-world projects, and advance your career

Pytorch for statisticians

A statistician's approach to the basics of pytorch

  • torch.randn
  • Compatibility with numpy (np.array <> torch.tensor)
  • torch.reshape
  • Various statistics on tensor
  • torch.mul
# Import torch and other required modules
import torch

torch.randn

Creates a tensor with the specified dimensions filled with values sampled form the standard normal distribution. (mean 0, variance 1)

Useful when you need a quick example, like in the following 4 functions.

# Using desired output size as input 

torch.randn(2)
tensor([ 0.2757, -0.5410])
size (int...)

a sequence of integers defining the shape of the output tensor. Can be a variable number of arguments or a collection like a list or tuple.