PyTorch Functions

Python APIs

In this post, we will learn about few of the various available creational PyTorch functions:

torch.as_tensor
torch.arange
torch.full & torch.full_like
torch.eye
torch.reshape

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

torch.as_tensor

Convert the data into a torch.Tensor. A torch.Tensor is a multi-dimensional matrix containing elements of a single data type.

# Example 1 - basic tensor
a = torch.tensor([[1, 2, 3, 4.]])
print(a)
b = np.array([1, 2, 3])
print(b)
c = torch.as_tensor(b)
c

tensor([[1., 2., 3., 4.]])
[1 2 3]

tensor([1, 2, 3])

As shown above, we have created a basic tensor and a numpy array. The numpy array can be converted into a tensor by using the torch.as_tensor method.