AI Notebooks - Getting started

Learn how to simply bootstrap an AI Notebook

Last updated 11th April, 2022.

Objective

The OVHcloud AI Notebooks service provides you Jupyter or VSCode notebooks, linked to CPU or GPU resources, without the hassle of installing or operating them. This guide will cover the creation of a new notebook.

Requirements

Definition

Notebooks are files which contain both computer code (e.g. python) and rich text elements (paragraph, equations, figures, links, etc…). Notebooks are both human-readable documents containing the analysis description and the results (figures, tables, etc..) as well as executable files which can be run to perform data analysis. It's vastly used across developer world, especially in the data and artificial intelligence fields.

OVHcloud AI Notebook is our solution for managed Jupyter or VSCode notebooks. You can launch your notebooks quickly with the CPUs and GPUs resources you need. You also get secure user access, simplified use of your data, and the most popular artificial intelligence frameworks (TensorFlow, PyTorch, Hugging Face, Scikit-learn, ...).

The advantage compared to doing your own setup is that everything is already installed for you, and that you pay only for your notebooks while they are running.

Each notebook is linked to a Public Cloud project and specifies hardware resources along with a machine learning framework and an editor among those available.

You can create notebooks with the OVHcloud Control Panel (see below in this tutorial) or use the ovhai CLI.

AI Notebooks also provide an easy way to access data from your Object Storage, you can read more about it here.

Launch your first AI Notebook

Log in to the OVHcloud Control Panel, go to the Public Cloud section and select the Public Cloud project concerned.

Access the administration UI for your OVHcloud AI Notebooks by clicking on AI Notebooks in the left-hand menu and click on Create a Notebook.

Create a Notebook 01

Give a name to your Notebook. Next, click on the Next button.

Create a Notebook 02

Choose the code editor you want. In this example, it's JupyterLab, but you can choose Visual Studio Code.

Both of them have their own pros and cons, but usually we can consider Jupyter as easier to user for a start.

Select JupyterLab, then click on the Next button.

Create a Notebook 03

Note the Edit this step link on the right up corner to edit the previous step of the wizard.

Choose the AI framework you want to use and click on the Next button.

Create a Notebook 04

As you can see a lot of fameworks are available. This guide uses one of the most famous, PyTorch. As you can guess all of these Frameworks are deployed as container based on images.

Next, select your privacy settings and click on the Next button.

Create a Notebook 05

Public access will expose your data and code to anyone getting the AI Notebook link. Be careful and don't use it with sensitive data.

Next, select a location for your new cluster.

Create a Notebook 06

You can adjust the amount of resources by clicking on the + or - buttons. Click on the Next button.

Create a Notebook 07

Next step is about storage options.

By default, you will have an ephemeral storage space (local storage) and in this step you can link Object Storage containers to your AI Notebook, and directly play with your data.

You can also download Git repositiories easily in your AI Notebook.

This step is optional, you can configure these options later.

Leave them blank and click on the Next button.

Create a Notebook 08

SSH public keys section is optional, click on the Next button.

Create a Notebook 09

At the end of the wizard, review your settings and click on the Create a Notebook button to launch the creation of your notebook.

Create a Notebook 10

Note at the bottom of the screen the equivalent of all these steps with the Command Line Interface (CLI) ovhai.

When your notebook is created, it will appear on your dashboard.

Create a Notebook 11

At this point your AI Notebook is created, you can test it with a simple Hello World as explained below.

Click on the JupyterLab link in the Access column in the array that lists all the created AI Notebooks.

Create a Notebook 12

If you have not already created a Public Cloud user account, follow this guide and then log in with your Public Cloud user account.

Click on the Connect button.

Create a Notebook 13

Next, click on Python 3 (ipykernel) button.

Create a Notebook 14

Next, enter the following code in your code section:

print("Hello World")

And run the code by pressing the ▶️ button:

Hello World

Your code is executed in your browser and will consume the CPU and GPU resources linked to your AI Notebook.

Create a Notebook 15

You can save your example by clicking in the sub menu Save of the File menu.

Stop the AI Notebook

Go back to the OVHcloud Control Panel. In the AI Notebook panel you can directly stop the desired notebooks.

Considerations

  • A notebook will run indefinitely until manual interruption, meaning that you will pay for it.
  • When you stop an AI Notebook, you release the compute resources, but we keep the data from your workspace. It will be billed at the price of OVHcloud Object storage.
  • Billing is per minute. Each started minute is due.

Notebook lifecycle

During its lifetime the AI Notebook will transition between the following statuses:

  • Billing starts once a notebook is Pending and ends when its status switches to Cancelling.
  • Only notebooks in states Pending and In service are included in the resource quota computation.
  • Pending: The AI Notebook is starting, and volumes are synchronized from the Object Storage.
  • In service: The AI Notebook is running and can be accessed from your browser.
  • Cancelling: The AI Notebook is still running, but an interruption order was received and RW volumes are uploaded to your Object Storage.
  • Stopped: The AI Notebook is stopped and RW volumes have been synchronized back to your Object Storage. Compute resources are released.
  • Deleted: The AI Notebook data is fully deleted, you don't pay anything.

Feedback

Please send us your questions, feedback and suggestions to improve the service:


Did you find this guide useful?

Please feel free to give any suggestions in order to improve this documentation.

Whether your feedback is about images, content, or structure, please share it, so that we can improve it together.

Your support requests will not be processed via this form. To do this, please use the "Create a ticket" form.

Thank you. Your feedback has been received.


These guides might also interest you...

OVHcloud Community

Access your community space. Ask questions, search for information, post content, and interact with other OVHcloud Community members.

Discuss with the OVHcloud community

In accordance with the 2006/112/CE Directive, modified on 01/01/2015, prices exclude VAT. VAT may vary according to the customer's country of residence.