Introduction to Deep Learning with PyTorch
Chapter 3: PyTorch Tensors
Installing PyTorch
PyTorch Tensors
In the previous chapters, all the variables manipulated in our gradient descent algorithm were of type float
.
To make all our operations vectorisable, we group our data into multi-dimensional matrices called tensors.
In this course, as in a high majority of courses provided in the Department of Computing, we will use PyTorch
as a Machine Learning framework.
Hence, all our data will be grouped into torch tensors (of type torch.Tensor
).
Installing PyTorch
Before implementing anything, let’s install torch
.
$ pip3 install torch torchvision numpy # numpy sometimes does not come with `torch`
If you want to do that in a virtual environment venv_pytorch
(recommended):
$ python3 -m venv venv_pytorch
$ source venv_pytorch/bin/activate
(venv_pytorch) $ pip install torch torchvision numpy