Scribd
Upload
26 views2 pages

1A PyTorch Installation

This document provides a step-by-step guide for installing PyTorch using Anaconda for Python 3.6. It includes instructions for downloading Anaconda, setting up a conda environment, installing necessary dependencies, and verifying the PyTorch installation. The guide also offers options for CPU and GPU installations, with recommendations based on user needs.

Uploaded by

Sachin Pradhan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
26 views2 pages

1A PyTorch Installation

This document provides a step-by-step guide for installing PyTorch using Anaconda for Python 3.6. It includes instructions for downloading Anaconda, setting up a conda environment, installing necessary dependencies, and verifying the PyTorch installation. The guide also offers options for CPU and GPU installations, with recommendations based on user needs.

Uploaded by

Sachin Pradhan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

1A_PyTorch_Installation

March 26, 2025

1 PyTorch Installation

A walkthrough to install PyTorch for Python 3.6. If you already have PyTorch on your machine
and feel confident in your installation, you can go ahead and skip this notebook.
For the purposes of this tutorial, I’m recommending using Anaconda as your Python distribution
and for managing your environments. If you’d rather use some other distribution or version of
Python and feel confident in your abilities, you may also choose to just follow the instructions here:
https://pytorch.org/get-started/locally/.

1.0.1 Install Anaconda

Download and install Anaconda. A more detailed version of the below steps specifically for Windows
(macOS and Linux instructions are similar) can be found here if you need additional assistance.
1. Go to the Anaconda install page and click on the Python 3.7 installer matching your operating
system (e.g. Windows, macOS, Linux) and system architecture (likely 64-bit X86).
2. Open the installer from your downloads folder and run it by double clicking on it.
3. Follow the installer’s instructions.
Note: there is an option at one point to add Anaconda to PATH. While the installer recommends
against this option (it’s possible it may interfere with your other software), we typically recommend
doing so, as it makes accessing Anaconda commands easier. The rest of this tutorial assumes that
this option has been selected.

1.0.2 Set-up a conda environment

1. Open a Command Prompt (Windows) or a terminal (macOS or Linux)


Note: If you chose not to add Anaconda to PATH during Anaconda installation above, you
will need to use the Anaconda Command Prompt instead.
2. Create a conda environment for PyTorch: Shell conda create -n pytorch
python=3.6
3. Activate your new environment:
conda activate pytorch

1
4. Install some supporting dependencies Shell conda install h5py imageio jupyter
matplotlib numpy tqdm
5. Install either the CPU version or GPU version of PyTorch:
CPU: (Recommended as sufficient for this class) “‘Shell # Windows/Linux conda install
pytorch torchvision cpuonly -c pytorch
# macOS conda install pytorch torchvision -c pytorch “‘
GPU: (You’ll also need to install some Nvidia software) “‘Shell # Windows/Linx, assuming
CUDA 10.1 conda install pytorch torchvision cudatoolkit=10.1 -c pytorch
# macOS # Trickier. May require installing from source. See PyTorch docs. “‘
Note: If you have GPUs available, you should use the GPU version for any serious research
or application, as it can be significantly faster. For the purposes of this demo though, the
CPU version is sufficent.

1.0.3 Verify PyTorch Installation

1. While in your pytorch environment, enter the Python shell with:


python
Verify that your python version says Python 3.6.[x].
2. Enter this program into your Python shell:
import torch
print(torch.__version__)
x = torch.rand(2, 3)
print(x)
Your command line should return something like:
1.3.1
tensor([[0.3651, 0.7264, 0.9823],
[0.5398, 0.2145, 0.3616]])
The version returned will depend on what the current stable version of PyTorch is when you
perform your install. As long as it’s 1.[x].[y], these notebooks should work fine. The numbers
in the tensor x will be random, so as long as the print returns a 2x3 tensor of random values,
you’re in good shape.
[ ]: import torch
print(torch.__version__)
x = torch.rand(2,3)
print(x)

You might also like