Hi! PipelineController has method: - add_pipeline_tags ( bool ) – (default: False) if True, add pipe: tag to all steps (Tasks) created by this pipeline. Can ...

Hi, I have such a usecase: I have different types of tasks that use GPU: task that uses 1 card task that uses 0.5 of card task that uses [other number] of ca...

Hi, are there available somewhere examples of testing in ClearML? For example unit tests that check if parameters are passed correctly to new tasks etc.?

Hi. When we construct a pipeline consisting of steps and when we are aborting a pipeline, the steps aren't aborting (they run until they are done). Is there ...

Hi! I am researching different MLOps libraries / platforms. I don't want to use platform as a service solutions. Could you suggest me what are the main diffe...

Hi, I am trying to use agent with a sample, very simple task. But it stucks and task does not finish. In UI in console I see what I pasted on image. Do you k...

Hi, is there a possibility to switch off the agent just after if finishes current task (not to fail the task, but to wait minimal time)?

I also have a general question. Most frequently, we work with our ml models not as a single script, but a project - multi level directory of files. How do we...

Hi, is there a possibility to use one GPU card with 2 agents concurrently (there are tasks that need only fraction of a card)

Hi, is there a way to abort task (not reset, not delete) from code?

Hi, I have a question about logs, when I am looking into apiserver logs, I have for example [2023-04-18 14:47:42,043] [9] [INFO] [clearml.service_repo] Retur...

Hi, I'm using a self hosted version of ClearML. I am querying experiments from UI and I am getting such error: Fetch experiments for selection failed Error 1...

Probably my question will be about something other than core concepts of ClearML, but something connected I think.... I want to schedule tasks on GPU and I w...

When we run some agents and then kill them, they remain in UI for quite a long time (even if they are don't exist) - it is like 5min. It there some way to ma...

Hi, I know that ClearML uses local changes for patching and running script. Can it also do the same with local commits?

When we run a task on GPU, we can access GPU monitoring. But can we access it from code? Usecase is: when we see that there is enough resources for some task...

Hi, do you know more or less how many workers can server work with. Have you make such stress-tests? The thing is that we have some power run agent on (for e...

I have question about a situation, when agent is running on a queue, but also we'll use clearml-agent execute --idHow will it work, a) will be run immediatel...

Hi, I have a case when I want to clone tasks and set some parameters for them. I noticed, that I can't pass numbers, only strings are possible there. When I'...

I would like to make a retry mechanism on tasks (if failed, then give some more chances to run correctly). When tasks is failed, I can't put it to the queue ...

Hi, by accident I've scheduled 1000 experiments. I would like to abort them. In the UI I can do "load more", but it loads 15 more. Is there any option to do ...

Hi, we have quite an unusual issue. We run some agents, we attach them to queue. They are doing the job (they are doing hyperparameter optimization), however...

Hi, can I run an ClearML Agent on multiple computers (on-premise)? Is there any example in the documentation on how to do that?

Tasks can be put in draft state - if we will execute: task.execute_remotely()it will not start Is there a way to do the same with pipeline (not run it immedi...

Hi, suppose, that I run Server on machine A and ClearML Agents on machines B_1, B_2, ....B_n . Is it required to open some ports on machines B_i ? I mean, is...

Hi, when we use Task.clone()we need to pass project id (not project name). Is there a way to get project id from project name somehow?

In UI such strange artifact is visible sometimes. After reloading page looks fine.

Hi, I think I've found a strange bug in scheduler. I would like to run some job every Monday at 15:40 UTC. So I run this at 15:35: scheduler.add_task( name='...

Hi, I am trying to use agent, but I have a problem. Execution of task stucks like this

Hi! In UI I see USER for each task. Where is this variable stored in the system. How to specify a different user when running a task?