I'm trying to run mober train with the --balanced_sources_src_adv option set to True and keep getting this error:
sklearn.utils._param_validation.InvalidParameterError: The 'classes' parameter of compute_class_weight must be an instance of 'numpy.ndarray'. Got [0,1,2,3] instead.
It seems the error is from 44 in mober/core/data_utils.py.
Line 40 should probably be
sorted_classes = np.sort(np.unique(class_series))
Also, from looking at the code, it seems like the --balanced_sources_ae flag is not actually implemented. If this is true, could it be added as a functional option?
I'm trying to run mober train with the --balanced_sources_src_adv option set to True and keep getting this error:
sklearn.utils._param_validation.InvalidParameterError: The 'classes' parameter of compute_class_weight must be an instance of 'numpy.ndarray'. Got [0,1,2,3] instead.
It seems the error is from 44 in mober/core/data_utils.py.
Line 40 should probably be
sorted_classes = np.sort(np.unique(class_series))
Also, from looking at the code, it seems like the --balanced_sources_ae flag is not actually implemented. If this is true, could it be added as a functional option?