ml-workspace
ml-workspace copied to clipboard
ml-tooling
Update?
Can we expect updates anytime soon? I really like to use ml-workspace. But since most of its software is outdated it becomes inconvinient to upgrade everything via pip / apt everytime I deploy a new container.
You can edit the Dockerfile with the updated versions of stuff (if a specific version is given) and then build the image. Then hopefully you will get updated version out of box. Unless some bugs occur with one of the updated libs/apps.
Thanks for jumping in with the helpful advice @MohamedElashri to help out 🙂
We have also started to update the libraries and tools in this branch; the light and minimal versions are buildable with that and they look good so far. Next step is updating & building the full version (see the runbook). We will post an update here as soon as we have pushed the new images.
@raethlein Do you intend to update either the R or Spark flavors as well? Definitely interested. Thank you for your work!
Does anyone here know how the tensorflow version is set? I was thinking of seeing if I could update to 2.7. Is it part of the miniconda version? (And ideally, I'd switch to miniforge.)
This issue is stale because it has been open 90 days with no activity. Remove stale label or comment or this will be closed in 14 days
![github-actions[bot] avatar](https://avatars.githubusercontent.com/in/15368?v=4 "Published by a pub.dev verified publisher"){kind=small}
This issue is stale because it has been open 90 days with no activity. Remove stale label or comment or this will be closed in 14 days
Can anyone provide on information as to if/when this will be addressed?
Up vote on this. Is there any update recently ? MlWorkspace is really a good tool for my usage
You should just use jupyterlab or code-server and install all extensions or libraries you need over there.
Its been too long since the last update that it might even represent secruity issues and you are missing out on new features because of old dependencies.
You should just use jupyterlab or code-server and install all extensions or libraries you need over there.
Its been too long since the last update that it might even represent secruity issues and you are missing out on new features because of old dependencies.
Better, you can update all tools and packages used in Dockerfile and keep using ml-workspace.
But ml-workspace requires node 14.
Mind sharing your Dockerfile if you think its so easy to update every dependency?