Basic Quickstart
Use this documentation to get a basic introduction to NVIDIA AI Workbench. For the full list of quickstarts, see Quickstart Guides.
In this quickstart, you use the AI Workbench desktop application to perform the following tasks:
Before you can complete the steps in this quickstart, you need the following:
NVIDIA AI Workbench is installed on your local system. For more information, see Install AI Workbench.
Start AI Workbench by double clicking the icon on your desktop, or by searching for the program and pressing enter. The icon has the NVIDIA logo in it.
AI Workbench starts and the My Locations page appears.
On the My Locations page, for Select a location, click Local.
AI Workbench connects to the location and the My Projects page appears.
On the My Projects page, in the Start a New Project box, click New Project.
The New Project dialog starts.
In the New Project dialog, do the following.
For Name, enter hello-world.
For Description, enter A Hello World example project.
For Local Path, accept the default path.
Click Next.
The NGC Catalog tab opens.
On the NGC Catalog tab, select Python Basic for the base container environment in your project, and then click Create.
NoteLater you can choose other base environments for your projects. For details about about each available container, see NVIDIA AI Workbench Containers.
Your new project opens and AI Workbench begins to build the container.
NoteDepending on your computer and network, the Python Basic environment should download and build in a short time. Other container environments that are larger and more complicated can take between 10 minutes and 15 minutes to download and build, depending on your computer and network.
While your project builds, you can do the following:
You can track the build progress in the status bar of the AI Workbench window.
You can see the logs for the build by clicking Building or Build Ready in the status bar.
After your project finishes building, Build Ready appears in the status bar.
(Optional) Click Environment, and then click Start Environment.
The container environment starts. If you skip this step, the container environment starts automatically when you do the next step.
Near the top of the AI Workbench window, click Open JupyterLab.
JupyterLab opens in your default browser ready for you to start working.
In JupyterLab, do the following.
Navigate to the code folder.
Create a new notebook by choosing Python 3 (ipykernel).
In the new notebook, enter
print("Hello World!")
.Click run to verify the code works correctly.
Save the notebook, and name the file hello-world.ipynb.
In JupyterLab, choose File, and then click Shut Down. Confirm shut down and close the browser tab.
In AI Workbench, wait until you see 0 apps running in the status bar.
Click Environment, and then click Stop Environment.
You can add code files to your project in the following ways:
You can drag and drop files into the folder directly in the AI Workbench interface. Typically, code goes in the code folder.
You can move files from your local file system into your project by using file browsers or the command line.
You can set up a mount to the host filesystem in AI Workbench from a source location to a target location inside the project container. For more information, see Mounts.
Each AI Workbench project is stored as a Git repository. When you are ready to save changes to your project, you commit and push the changes to your Git repository. For more information, see Connect to Git.
To create a commit that contains your changes, click Commit near the top of the AI Workbench window.
The Create a commit window appears and contains a detailed list of the changes to your project.
In the Create a commit window, edit the commit information, and then click Commit.
To publish your changes to a Git server, click Publish near the top of the AI Workbench window.
The Publish Project window appears.
In the Publish Project window, edit the Git information, and then click Publish.
Open your Git repository manager and verify that you see your project.
Do a Quickstart
Explore the NVIDIA Developer Forum
Learn the Concepts