Martin.B

Hi @agunapal , this is somewhat similar to #8 #9 In short when executing your code "manually" inside a python (pip) environment , the *trains* package records all the environment...

@jan-zajac confirmed, for some reason the new poetry version cannot be directed to a target folder :( I think this is what we need to do: https://python-poetry.org/docs/configuration/#virtualenvscreate-boolean Let me play...

@LanceKnight can you see the command line argument in the "Args" section under the "Configuration" tab ?

Ohh I think the confusion here is that `cleaml` actually will catch your ArgumentParser not directly the command line arguments. This allows it to store all the defaults along side...

Hi @klekass > I am using 'click' instead of argparse for my CLI, May I suggest you add a github issue to [here](https://github.com/allegroai/clearml/issues) and suggest `click` integration? wdyt? > Using...

Hi @bzamecnik > UPDATE: I found that a copy of the whole ~/.ssh directory and also a cache of the git repo from the host machine is mounted into the...

> (maybe because local .ssh has the very strict - recommended - permissions) Yes, it has :) https://stackoverflow.com/questions/9270734/ssh-permissions-are-too-open-error > Any possibility of passing the .pub file as an option to...

Thanks @AzaelCicero ! > I would expect it to run command with NVIDIA_VISIBLE_DEVICES set to correct value, Yes one would expect that but life is strange ;) The way the...

@AzaelCicero I see the issue `--privileged` will cause the docker to see all the GPU's, see issue [here](https://github.com/NVIDIA/nvidia-docker/issues/764) You can quickly verify: `docker run --gpus device=1 -it nvidia/cuda:11.2.0-devel bash` then...

@AzaelCicero I think you are right, even though not "traditional" setup, I think `clearml-agent` should properly handle it. In order for the agent to pass the correct "NVIDIA_VISIBLE_DEVICES" (i.e. understanding...