My favorite pytorch functions

Pytorch is python library which extends the functionality of numpy to building regression models. A functions that I picked in order to share what I have learnt.

tensor.stack
tensor.randn
tensor.abs
tensor.add
tensor.var

Before we begin, let's install and import PyTorch

# Import torch and other required modules
import torch
import numpy as np

Function 1 - torch.stack

Concatenates a sequence of tensors along a new dimension.

x=np.array([[3, 5],[7,8]])
print(type(x))
print(x.dtype)

<class 'numpy.ndarray'>
int64

z=torch.tensor([[4,5],[9,1]])
w=torch.tensor([[8,2],[0,2]])
print(z.shape)

torch.Size([2, 2])