Hi, I'm conceptually struggling to understand how to get my models to be saved permanently by my self-hosted ClearML server. As it stands now, when I train a...

After resetting and enqueueing a task, the pip installation procedure in a remote agent tries to install packages different from packages that were installed...

Hi, I'd very much like a feature to support clearml agent stopping specified by the exact bringup command: when trying for example to bring down a gpu agent ...

Hi, when running a clearml agent with services mode, is there a way to limit the number of concurrent services crunning?

Hey, is there a way to run tasks not on ClearML agent but rather as a k8s job directly? Thanks!

Hi, I have a question regarding the aws_autoscaler: It usually takes ~hours to get a GPU instance nowadays. I was thinking, it would be much more interesting...

Hi guys! Is there a way to tell an agent to run a task in an existing venv (without creating a new one)?

I'm working on lsf cluster and wonder what is the easiest way to use the queue without having permanent agent or submitting new agents to the lsf system peri...

Hi, I'm trying to clone and queue experiments for running them on my workers. I am able to successfully clone and queue the task, but seems like the task doe...

For some reason the agent is now trying to use Python 2.7 all of a sudden, any idea why? 🤔 I see the following in the task's console output: New python exec...

Another question regarding a curiousity. Given a task is sent to an Agent to run and it has a specific version of tensorflow/pytorch, which requires a specif...

Hi everyone, we train our ML models using the AWS autoscaler on g4dn instances. We currently have a 24 vCPU limit for G type instances in eu-west. I'm trying...

Hi , just updating that we've found an issue related to the current ClearML Agent requirements and the recent release of pathlib2 v2.3.7 (due to conflicting ...

Hi, I am trying to use clearml agents with docker. I send a task into a queue, an agent grabs it and starts it and everything seems fine - but the task gets ...

Hello, I am trying out ClearML and started an experiment with Tensorflow Object Detection API. I got stuck with an issue regarding the local package " object...

Hey, I’m running jobs remotely on an agent but one of my packages uses numpy 1.21.2 which needs python3.7 and above but the python version on the agent is 3....

We're trying to use the AWS autoscaler and have managed to get it up and running with spinning up instances. However, it does not seem to pull any of the tas...

What privileges/IAM role would the AWS autoscaler need?

Hi all, Are there any examples in the official (or external) documentation where I can see how ClearML and Docker work together? This comes related to anothe...

Is there a way to set precedence on package managers? If we set an agent to use poetry it will ignore local packages, even if I use the force_requirements_en...

hi, i want clearml to not install packages from my pycharm environment, but activate an environment i have locally on the computer the agent is installed on,...

Is there any way to stop all clearml agent workers on a machine or stop workers from the clearml ui?

Different question. How can I pass PYTHONPATH env variable to a task, run by agent (so python can find classes inside m subdirectories)?

Set GIT user/pass credentials (if user/pass are set, GIT protocol will be set to https) # leave blank for GIT SSH credentials (set force_git_ssh_protocol=tru...

Question about pipelines - so the default for pipeline tasks that are executed remotely is to execute on the services queue. Assuming I want only ONE agent l...

Hi, did anyone experiment with running on the aws autoscaler on spots and knows whether there is configuration for retry policy when spot get evacuated mid-job?

I'm trying to understand how clearml serving works and trying to set it up. I have an agent listening to the serving queue and I'm trying to set up clearml s...

Hey. Seeing a strange error when trying to install a private python package when the agent executes the task. We use AWS CodeArtifact as the private PyPi rep...

Hello, I have a server with several GPUs that I wish to use to automatically train models. ClearML seems like the perfect tool for my use case but I am confu...

Does anyone know why I keep getting that error? clearml_agent: ERROR: ERROR: package manager "conda" selected, but 'conda' executable could not be located I ...