Hi all,
I am trying to expand our training infrastructure using Clearml-agents, I have a train_example script that I want an agent to be able to pick up and execute. However, the configuration of the hyperparameters of this train script is in the Task, which I can't seem to retrieve easily. The examples on the website all mention Task.get_task(project_name, task_name), but these variables are different from one execution to the next (and also defined in the configuration) and when I call Task.current_task() in my script it comes up None . The way that we currently retrieve our task is Task.init(continue_last_task=True) but this seems like a roundabout way of doing it. Has anybody found a more elegant solution to this?