Note for students: Configuring a deep learning environment, especially with GPU support, is a complex task. In addition, the procedures and scripts in this repository are under active development and have only been tested on my personal machine.
I recommended scheduling a one-on-one session with me to complete the setup process. This will allow me to provide real-time guidance and troubleshooting support, which can be invaluable when dealing with potential issues that may arise during installation. In addition, doing it right the first time is much more efficient than trying to fix a broken setup later on.
To complete course activities (i.e., assignments and projects), you will need a properly configured deep learning environment. Below are the options available to you, along with their advantages and disadvantages. Please review these options and choose the one that best fits your needs and resources.
| # | Platform | Description | Advantages | Disadvantages |
|---|---|---|---|---|
| 1 | Codio | A pre-built, cloud-based, Linux virtual machine accessible through a web browser from our course site. | - No setup required - SNHU Tech Support |
- Time-out/loss of work - Performance limitations, especially on weekends - Less control over the environment - Requires a stable internet connection |
| 2 | Local Setup | Setting up a deep learning environment on your personal computer. | - Full control - Better performance for training models - Can work offline - Great skill to develop |
- Requires more time and effort to set up - May encounter compatibility issues - Requires a reasonably powerful computer |
| 3 | Web | Using cloud platforms like Kaggle notebooks or Google Colab for development and training. | - No local setup required - Access to powerful hardware (e.g., GPUs) - Easy to share notebooks and collaborate |
- Requires a stable internet connection - May have usage limits or costs associated with heavy usage - Less control over the environment |
Note. The Codio VM is the SNHU-recommended option for this course. However, setting up a local environment is a valuable skill and may provide better performance for training deep learning models. In addition, the MSCS program requires students to have a computer with a GPU—might as well put it to work.
If you choose to set up a local environment, please follow the instructions in the next section.
Follow the guide for your operating platform to set up your deep learning environment. Each guide will provide step-by-step instructions to get you up and running.
| # | Platform | Processor | Setup Instructions | Status |
|---|---|---|---|---|
| 1 | Codio | CPU-only | codio/README.md | Complete except for student process and view of Codio dashboard page. Request a student to take some screenshots. |
| 2 | Win+WSL | CPU+GPU | wsl/README.md | complete except for VS Code + extensions |
| 3 | Kaggle / Colab | CPU+GPU | web/README.md | not started |
| 4 | Linux | CPU+GPU | nix/README.md | not started |
| 5 | Win-only | CPU-only | win/README.md | not started |
| 6 | macOS | CPU-only | mac/README.md | not started |
Note. Platforms are listed in development order. If you have need for a platform that is not yet available, please let us know by opening an issue in this repository or posting in the course General Questions discussion board.
If you are CS 580 student, open an issue in this repository or post in the course General Questions discussion board.
This supplemental course resource contains content created with the help of AI tools. AI-generated content has been thoroughly vetted by course instructor. Revisions have been made to ensure accuracy and clarity.