ValueError: The number of classes has to be greater than one (python)

Exactly. your last column (label) has only one type (Classification). you should have at least two. For example; if your label is to decide either you have to offload or not, the label column should have offload and not-offload or (0 or 1).

You probably have only one unique class label in the training set present. As the error messages noted, you need to have at least two unique classes in the dataset. E.g., you can run np.unique(y) to see what the unique class labels in your dataset are.