Johannes Gasteiger, né Klicpera

I get the exact same error. When two projects are open at once, the path is interpreted as a concatenation of the two projects' paths. So, when I try to...

Thank you for the suggestion! I'm afraid I don't have the time to implement this right away. But maybe at some later point. In the meantime, maybe some of our...

In #79 Daniel mentioned that this should allow overriding Slurm parameters when _starting_ an experiment, which is not addressed by this design. However, I don't think it's ideal to have...

Yes, good point! Let's keep the PRs small and manageable. update could also break our assumption that every batch has consistent Slurm parameters. Creating batch arrays and sharing GPUs both...

Thank you so much for spending the time on this and posting such a great issue description! So the experiments are only set to KILLED in the MongoDB, while they...

That's an easy fix then, since the Slurm states are defined in the settings. You just need to adjust the running states. Either in the repo's settings [here](https://github.com/TUM-DAML/seml/blob/master/seml/settings.py#L57) or do...

Indeed, the only documentation we currently have is the examples folder. Proper documentation has been on our to-do list ever since we started with SEML, but we never got around...

I like the idea, but find the term "group" a little ambiguous. There are so many things that a "group" could refer to. Is there a better term? Something like...

We're using UUIDs in several other places in SEML, though. And relying on their defining uniqueness simplifies the code in all of those places.

I think I prefer `bundle` to `group`. The picture it invokes is better. Also, with `bundle` the user thinks this is something special and looks it up, instead of just...