Hi, I’m trying to bypass Task’s “installed packages” and require an installation of my repo’s requirements.txt . I’ve set agent.package_manager.force_repo_re...

Since PipelineDecorator automatically starts the task for you, is there any way to specify arguments to Task.init in the task created for a function decorate...

Hi! Any idea why clearml fails to detect iteration reporting? ClearML Monitor: Could not detect iteration reporting, falling back to iterations as seconds-fr...

hello, is there a way for the clearml agents to check for repository changes recursively? eg, our source is broken in serveral different submodules, clearml'...

Is it possible to ask an agent to use a specified existing python environment instead of building one from scratch?

Hi everyone, is it possible to show the upload progress of artificats? E.g. I use torch.save to store some very large model, so it hangs forever when it uplo...

I have no prior DevOps experience. I've been able to set up a simple continuous training setup using clearml. I wanted to ask what should I learn which would...

Hello, I'd like some help debugging why my ClearML server is crashing. I'm currently running my server inside GCP, and I'm not being able to access it anymor...

I have a question regarding the deletion of archived experiments. Some of them can't be deleted and the error message is General data error (TransportError(5...

Hi all, I'm getting this "ERROR - Action failed <500/100: events.add_batch/v1.0" in my task log, can any one help diagnose \ solve this issue? what does it a...

we are facing performance issues of our self-hosted ClearML server Looking at the cpu utilization \ memory \ networking we couldn't identify a bottleneck we ...

I have another small technical question, I am trying to see the workers status programatically using the folowing: from clearml.backend_api.session.client im...

I have a question about the clean up script. The cleanup service can remove model checkpoints that are saved somewhere on disk. However the cleanup service i...

When ClearML converts a PipelineDecorator.component decorated function to script code, I have noticed that indexing syntax like A[:, 0] is rewritten as A[(:,...

How to use task_overrides in PipelineController.add_step method to set latest commit in some branch?

How come PipelineController has changed in a very backwards-incompatible fashion but we are still on the same major version? O_O Also the documentation is no...

What's the difference between http://allegro.ai/clearml/docs and http://clear.ml/docs ? Which one is the real documentation? I need to look up something very...

Hi! I was wondering if ClearML offers any solutions for the following situation. I have created a queue (say Q) for model training tasks, and two agents list...

Hi There everyone! We're working on some better UX for triggering and scheduling functions and would love to pick your brains 🙂 If anyone is using it \ inte...

New RC(1.1.2rc0) version available! 🎉 Change set: ClearML Data - Upload dataset now supports chunksize, for multi-part upload/download (useful with large da...

Hi all, I have a Live Webinar coming up on Feature Engineering with KubeFlow Feast on Oct 7th, 2021 | 02:00 PM CST... https://www.royalcyber.com/webinar/feat...

Avoiding http://Clear.ml glue code spaghetti - community best practices? Say I have training pipeline : Task 1 - data preprocessing -> create a dataset artif...

A suggestion. Sometimes newcomers that join an existing project that uses ClearML forget to configure their ClearML for the organization's server resulting i...

Hi, I am having difficulties when using the Dataset functionality. I am trying to create a dataset with the following simple code: from clearml import Task, ...

Hey, ClearML team! When can we expect an updated roadmap? Last one is from august https://docs.google.com/document/d/1QlPiDO2EzDq_HRvuVhYwPh5F_IsjgmmRTlPNdHi...

Hi there, is it possible to configure the clearml-agent to run some commands before running each experiment it launches? Eg. echo "test" > "test.txt" && <-- ...

I have a question. If I do a training on a cloud instance for example the models are saved there first. This is the path that is also logged in the clearml a...

Can I prevent torch.save()from automatically uploading the model and use task.update_output_model(model_path=best_model_path)at the end to upload the best mo...

If I'm using a drive mapping to save files is there any easy method/hack that would allow me having different base mappings on different machines?

I'm a little confused as to how force_requirements_env_freeze works when no requirements file is supplied. Is it supposed to store the full reqs of the envir...