[Data Science dev containers] Do you think this setup can work with DevPod?

[Analect]

Thanks for your work on these devcontainers. I was wondering if you have experimented with using the open-source DevPod, which can then be used in conjunction with a provider like Civo that has GPU's with CUDA available. Other than cloud-providers, you would also be able to add a SSH and kubernetes provider (an existing cluster). Thanks.

[benz0li]

I was wondering if you have experimented with using the open-source DevPod

Yes, I did. For local use I recommend VS Code, though. Please follow the instructions at Developing inside a Container.

which can then be used in conjunction with a provider like Civo that has GPU's with CUDA available. Other than cloud-providers, you would also be able to add a SSH and kubernetes provider (an existing cluster).

For remote use I recommend a deployment based on Træfik, GitLab and Jupyter.
👉 A self-hosted, full-featured Data Science software stack – for free.

🔬 Check it out online at https://demo.jupyter.b-data.ch

[benz0li]

@Analect Does this answer your question?

[Analect]

Thanks @benz0li ... I had missed this back in June. I have previously leveraged your traefik/gitlab/jupyter setup.