Is there anyway to store the DedupeDisagreementLearner object efficiently?

[seperman]

Hello,
Thanks for the great library.

1. I want to dump the DedupeDisagreementLearner object for fast re-loading if the console_label step throws an exception.
   The pickle dump of DedupeDisagreementLearner can get quite big. My goal is to store the parts of it that are slow to calculate. Any suggestions?

2. As a part of the dump, I am including the following objects.

               pickle.dump(self.data_model, the_file)
               pickle.dump(self.classifier, the_file)
               pickle.dump(self.active_learner, the_file)

   However I still get an error when loading the dump and trying to run the console_label:

   {'Site name': 'el hogar del nino fcch-blanca alatorre', 'Address': '5738 s kenneth ave', 'Zip': '60623', 'Phone': '5812872'}\
   is not known to to the active learner. Make sure all `labeled_pairs` are in the data or training file of the `prepare_training()` method
   

   It seems some other state other than self.data_model and self.classifier and self.active_learner needs to be dumped too. Am I correct?

Thank you