This MR adds two things:
- Make the chunk size of the model generators configurable as input parameter.
- Ensure that the number of jets (which is loaded from the config file) has the correct
int. Because if you define it as
1e5, it is loaded from yaml as a float ..