Hi there, I am looking for a way to use the hyperparameter optimization tool from Clear ML, for different algorithms I build. The idea is, that I can use a command line to specify which algorithm I want to optimize, and let clear ML do the rest. The issue is, that apparently the hyperparamter optimization of clear ml works in a way, that it calls the function where the task was created to run the training. So, if I have two function, one for the training and one for the optimization, I cannot call them in the same file, because after the training function, the optimize function would be called everytime, that leads to a lot of chaos, so my learning was that I have to separate the task from the optimization. However, if I want to use, for example, argparse to create the training, then I would need to pass in the arguments everytime this function gets called from the optimizer, which is not possible fully auomated. So i am kind of stuck how to make my idea work. Any ideas?