Skip to main content

Bring Your Own Container (BYOC).

info

This is an advanced feature, if you are getting started with DL it would be easier to choose one of the frameworks - PyTorch, TensorFlow or FastAI.

You can Bring Your Own Container - BYOC on Jarvislabs.ai.

Jarvislabs.ai BYOC

To launch an instance from public image follow the below steps.

  1. Choose BYOC under framework
  2. Enter a public image name like pytorch/pytorch. (Organization/repository) in the docker field.
  3. Ensure you have configured your SSH key in the API keys section.

Once the instance is launched, right-click on the instance to get the SSH string and connect to the instance. The ssh string would look like this

ssh -p port_number1 root@ssha.jarvislabs.ai -L port_number2:localhost:8888

Connecting to a jupyterlab inside the instance​

Install jupyterlab in the newly created instance.

pip install jupyterlab

danger

This step could change based on the image you choose. Some docker images can have Jupyter installed, or you may need to activate a conda environment. So look at the specific docker image docs, on how it is configured to use.

Starting JupyterLab​

jupyter lab --allow-root

Open a browser, and enter the below URL

http://localhost:port_number2/
info

Copy the port number from the ssh URL and replace the port number above. Use the token printed in the terminal while starting jupyter lab.

The feature is in beta, in case you face any challenges, please reach out to us. We will be happy to adapt and make the process smoother.

tip

Nvidia provides a variety of optimized NGC Containers, you can choose the required container from here.