Thanks!

Thanks. That would mean I cannot abort the training process, but have to wait until its finished to call the upload_artifact? Is there a reason why it's not logging all the generated files?

I could get rid of the error message not by adding

!pip install jupyter

But now the task starts and never finishes. I think it's because it starts a Jupyter Kernel in ClearML instead of just executing the code in my Notebook. Is there any configuration to tell ClearML it should just execute the code of the notebook remotely?

I executed the following code in a notebook:

from clearml import Task
task = Task.init(project_name='Test', task_name="TEST NOTEBOOK")
task.execute_remotely(queue_name="RTX A6000")
print("HELLO")

I ran the notebook as a standalone file. I don't have the exact start command, but it runs in docker mode on ubuntu 22.04 using the standard image. Is there anything to consider which might be related to the described issue?

I created a ticket for the issue with a more detailed explanation and log file: None

After reading the docs, is execute_remotely supported to be run from a notebook?