Hey all! Great job on ClearML by the way 🙂 I’m currently exploring whether it could be a fit for me and my team and I have a few questions: If I want to sen...

Is there an example of how to use standalone mode? As in, if the packages, venv and git repo will be provided by the user, will they still get tracked in the...

There’s a typo in this repo name: https://github.com/allegroai/nvidia-cleamrl-integration

I’m running the clearml agent in standalone mode but it still installs the virtual env, is that the expected behavior? clearml-agent daemon --queue venv_queu...

There is some specificity with the way we setup our environment at my company that prevents me from using the full features of clearml-agent and forces me to...

I see that in the default setup, this command is part of the docker bash setup script: chown -R root /root/.cache/pip Does the agent assume it will be runnin...

Also, since this Slack workspace only allows searching up to 10,000 messages, useful information will be lost overtime. It’s been very useful for me personal...

If we’re using the same git repo over and over for almost all jobs, is it possible to have the agents keep a local version and only download the diff of the ...

I’m getting these errors when using agent in docker mode -bash: /etc/apt/apt.conf.d/docker-clean: Permission denied chown: cannot access '/root/.cache/pip': ...

Do you have some documentation on what https://app.community.clear.ml/applications are useful for and how to use them? Are they going to be available for sel...

I’m trying to use task.execute_remotely but the repo I’m trying to use is not being identified correctly. I have it setup to git pull from a master repo and ...

Is it possible to give the agent access to install private pip packages (needs to be installed from the repo)? ERROR: Could not find a version that satisfies...

Another issue is the agent uses Python 2 for some reason even though locally I’m using python 3 and the agent is supposed to use a python 3 venv. $ /home/smj...

When scheduling a job for remote execution, is it possible to decide whether to use the repo’s formal requirements sheet vs the currently installed venv libr...

Related to my above thread, I think there is some more customization that needs to be supported to use docker mode properly for different use-cases. It seems...

Can someone please explain the difference between clearml-agent-services (that comes with the official docker compose) and clearml-agent deamon?

Is it possible to interact with the job queues directly from the python SDK?