4

I would like to run a Jupyter notebook from a docker image on GCP. Specifically I would like to run the one from kaggle/python. On my local machine I can get this working with docker pull kaggle/python and then the following docker cli command:

docker run -v $PWD:/tmp/working -w=/tmp/working -p 8888:8888 --rm -it kaggle/python jupyter notebook --no-browser --ip="0.0.0.0" --notebook-dir=/tmp/working/src

I do not want to use a gcloud cli command. How do I map this command to the GUI in GCP? I wan't to run this on Google Containerized OS:

enter image description here

Jonathan
  • 1,287
  • 14
  • 17
  • The GCP console works through API calls, as does the `gcloud` commands. Other than using `gcloud` to interact with the Compute Engine API, are through various programming language's [client library](https://cloud.google.com/compute/docs/api/libraries), & the [REST API](https://cloud.google.com/compute/docs/reference/rest/v1/instances/insert). See [here](https://cloud.google.com/compute/docs/apis) for more details about the compute engine API. You can also try using GKE to deploy containers. Is this what you're looking for? What are you trying to accomplish? – Sunny J Nov 23 '18 at 03:02

1 Answers1

1

You can now use docker containers to your custom notebook environment on GCP's AI Platform Notebooks. That might serve your means

They offer a base container that you can modify to fit your needs and then use as your active notebook environment

You can find more info about them here: https://cloud.google.com/ml-engine/docs/using-containers

Zain Rizvi
  • 23,586
  • 22
  • 91
  • 133