I have an ML project that is not on git. It is separated into several files:
I can monitor it easily through clearML. Everything works very well.
Now I am implementing a class for hyperparameter optimization.
I implemented the whole class, but when I execute it, passing the id of the task I want to base it on, it does not find the files (example: global_config.py).
I understand this because, as I understand it, it only monitored "training.py" where Task.init is.
Is there any way to link these other files to training.py? So it understands that it needs these files to run training.py?
Is git the solution?
Note: when running optimizer.start_locally(), this error occurs:
Traceback (most recent call last):
File "C:\Users\MATEUS~1.CAS\AppData\Local\Temp\tmpn75k0lbv.py", line 22, in <module>