Installing PyTorch GPU on Mac M1/M2 Using Conda

In this tutorial, we will walk you through the process of installing PyTorch with GPU support on your Mac M1 or M2 using the Conda package manager. PyTorch is an open-source machine learning library that is widely used for building deep learning models. By installing PyTorch with GPU support, you can take advantage of the computational power of your Mac’s GPU for faster model training and inference.

Before we begin, please make sure that you have installed the latest version of Conda on your Mac. If you haven’t already installed Conda, you can do so by following the official installation instructions on the Conda website.

Step 1: Create a new Conda environment
The first step is to create a new Conda environment for PyTorch installation. Open a terminal window and run the following command to create a new Conda environment named ‘pytorch-gpu’:

conda create -n pytorch-gpu python=3.8

Activate the new Conda environment by running the following command:

conda activate pytorch-gpu

Step 2: Install PyTorch with GPU support
Next, we will install PyTorch with GPU support in the newly created Conda environment. To do this, run the following command:

conda install pytorch torchvision torchaudio cudatoolkit=11.1 -c pytorch

This command will install PyTorch, torchvision, and torchaudio with GPU support using CUDA toolkit version 11.1. This version is compatible with the latest Mac M1 and M2 GPUs.

Step 3: Test PyTorch installation
After the installation is complete, you can test if PyTorch is working correctly with GPU support. Open a Python shell in the Conda environment by running the following command:

python

In the Python shell, import PyTorch and check if it is using the GPU by running the following code:

import torch
print(torch.cuda.is_available())

If the output of the above code is True, then PyTorch is successfully using the GPU for computation.

Step 4: Install additional packages (optional)
If you plan to use PyTorch for specific tasks such as natural language processing or computer vision, you may need to install additional packages. You can install these packages using Conda or pip, depending on your requirements.

For example, to install the Transformers library for NLP tasks, you can run the following command:

conda install -c huggingface transformers

Step 5: Save the Conda environment
Once you have installed all the necessary packages, you can save the Conda environment to a YAML file for future use. Run the following command to save the environment:

conda env export > environment.yaml

To activate the environment later, you can use the following command:

conda env create -f environment.yaml

Congratulations! You have successfully installed PyTorch with GPU support on your Mac M1 or M2 using the Conda package manager. You can now start building and training deep learning models with PyTorch on your Mac’s GPU. If you encounter any issues during the installation process, please refer to the official PyTorch documentation or seek help from the PyTorch community.