Hey there, since a bit I often find experiments being stuck while training a model. It seems to happen randomly and I could not find a reproducible scenario ...

In my git repo, I have a setup.py , how would I run pip install -e . rather than using --packages or --requirements

Hi everyone, I saw a new entry in docker-compose: async_delete Can someone tell me, what this is for (there seems to be not documentation or pull request) an...

Second question: I want to run clearml-server without fileserver. However, it seems like the webserver has some dependency on it (I get some nginx error). Am...

Hi everyone, I’m getting an error during model upload to S3. The error shows up in the console like below and I don’t see any uploaded objects in S3: 2022-10...

I have noticed that for some people the repository discovery stores github repo in the form: https://github.com/ ... while for others git@github.com:... Inte...

I have a general question about this part in dynamic GPU allocation. If for example I have a machine that has 8 GPUs and I have 3 queues: Queue1 will take 3G...

i think there is some bug with clearml==1.7.1. I’m working with pipelines and after updating to 1.7.1 all the kwargs are passed as None to the steps, when do...

In pipelines I've found that empty lists don't work as I would expect them to work. For example, this will work fine: from clearml.automation.controller impo...

Once I have a model successfully saved in ClearML, how can I instantiate it in a notebook? I have this for artifacts: def get_artifact(task_id, artifact_name...

There is a problem starting from clearml 1.7.0 with python-fire from clearml import Task import fire def check(first): print(first) if __name__ == '__main__'...

How can I figure out from the dataset object which task it created? Or better: I have a task_id how can I figure out which datasets were created by this expe...

Hello, I am trying to install the open-source server locally on my mac using the docker-compose file from repo. Everything is fine but the agent-services con...

For some reason I can't delete a pipeline projet, the deletion is running indefinitely. Is there a way to force the deletion of a project via the APIClient?

Hey all -- I'm fairly new to this but, as of today, my required packages aren't being recognized in cloned runs and they are repeatedly failing. Has anyone h...

From what I understand ClearML should be auto logging my model, how ever this isn't happening. I have clear ml installed via docker on a local server, and am...

Hi all, does anyone know how I can use python to load the configuration of a saved experiment/model using its id?

Hey, it is said in the pipeline decorator example that requirements for executing a pipeline component is inferred from the imports inside the component func...

How does the Preprocess class exactly work? I need to pass information to it but it's _ init _ function doesn't have any parameters. How can I load an artifa...

I deployed a model with: clearml-serving \ --id 875de894b62144a7949e471532728809 \ model add \ --engine "xgboost" \ --endpoint "best_model" \ --preprocess "t...

(the payload is not the correct form, can that be a problem? I'd rather figure this out before I recreate the rather convoluted data structure that needs to ...

Hey currently trying to run a pipeline locally to test a pipeline component with PipelineDecorator.run_locally() , first try returned a random pandas error, ...

I am completely stuck with the serving. I did the custom example. I see the endpoint in clearml-serving -id ... model list and it replies {"detail":"Error pr...

What exactly triggers the "automagic" logging of the model and weights? I've pulled my simple test project out of jupyter lab and the same problem still exis...

anyone setup pipeline decorator for training huggingface transformer model??

How can I remove a service with clearml-serving?

{"detail":"Error processing request: Error: Failed loading preprocess code for 'py_code_best_model': [Errno 2] No such file or directory: '/root/.clearml/cac...

Heya, trying to setup a GCP autoscaler for general-purpose CPU instances ( e2-standard-4 ) but I get into that error: googleapiclient.errors.HttpError: And I...

Hello, The problem: clearml UI (and service in general, e.g. task logging) is unreachable via the VPN-internal IP of the machine it was deployed on. Was reac...

Hey guys! I'm saving my datasets on our self-hosted ClearML file-server. However, when deleting them through the UI, it doesn't delete them on the server. I ...