Thanks Martin 🙂

AgitatedDove14 Quite hard for me to try this right now. but I've validated that the relevant code segments are untouched between the versions. (at least current master branch at the ClearML repo)

Nevermind, you can find it in the apiserver.conf

I've sorted this out. All I needed was to add them to a queue so they would be visible.

Nope, only its name

I see haha

If I change the file at the entry point (let's say, I delete all of its content), how will trains behave when I try to clone and execute such task?

TimelyPenguin76 it didn't help 😞

If this could be so, I'd be happy to have this as a feature, this really impacts my pipeline flow.

That should do the trick, thanks 🙂

You can try copying all the contents of requirements.txt to the installed packages tab in the trains dashboard of your experiment (in the UI)

My root folder is applicable to my user only. I wish to use a shared trains.conf file, so the trains_config_file can't point to ~/trains.conf sadly

So I'd guess they would inherit my user as well

I think I know what happens TimelyPenguin76
Could it be that trains automatically logs these images to plots?
Because when I removed the report_media/report_image the images were still logged into plots

SuccessfulKoala55 I found the temp files, they contain the supposedly worker id, which seems just fine

otherwise if you empty the installed packages and the requirements.txt is in one of the parents folder of the files that ran trains should detect it automatically

Is there a way to set this via a config file? like the docker compose yml?

how could I configure this in the docker compose?
Edit: the trains-agent points to a different trains.conf config as I wis., I want the dev environment to point to a different location trains.conf as well

Yeah I understand that. But since overriding parameters of pre executed Tasks is possible, I was wondering if I could change the commit id to the current one as well.
What do you mean by execute remotely? (I didn't really understand this one from the docs)

Could it be because it's running from a draft on an agent?

By the way, what about enabling closing workers from the UI? 🙂

Sure

For me

Since my servers have a shared file system, the init process tells me that the configuration file already exists. Can I tell it to place it in another location? GrumpyPenguin23

Oof, if all I have is a project bame to set? (Which could be a non existing project as well)

It what it does to my understanding

Great, I will try that 🙂

Furthermore, let's say I have 6 GPUs on a machine, and I'd like trains to treat this machine as 2 workers (gpus 0-2, 3-5), is there a way to do that?

Oh

I think it should be treated as failed, I am truly not convinced as why aborting a task should be anything beside a user terminating an unwanted behavior of the task (be it bug, running with wrong config, task getting stuck etc..)