Dedupe - AttributeError: 'NoneType' object has no attribute 'indexAll'

Question

I'm using dedupe library and everything works fine until training data is used for dedupe but while calculating the threshold with the same data set it gives the following error:

deduper.threshold(data_d, recall_weight=2)

AttributeError: 'NoneType' object has no attribute 'indexAll'

My data set:

{23347895: {'Address': '100 Kanoki street drive',
  'Id': '23347895',
  'Name': 'kanoki',
  'Unname': 'brunswick',
  'no': '1',
  'zip': '41092'},
 2543294: {'Address': '1005 Main Street',
  'Id': '2543294',
  'Name': 'kushal gore',
  'Unname': 'brunswick',
  'no': '2',
  'zip': '41093'}