b2 = si.block_cdist(synth, mpg, ['cylinders', 'model_year'], metric='euclidean')
/home/maxghenis/miniconda3/lib/python3.6/site-packages/pandas/core/reshape/merge.py:963: UserWarning: You are merging on int and float columns where the float values are not equal to their int representation
'representation', UserWarning)
This is probably due to rf_synth synthesizing model_year as a float (that's the dtype) where it's an int in the original data. This should then go away if warning when records are dropped because they're not part of a block.
In the
mpg
example:This is probably due to
rf_synth
synthesizingmodel_year
as a float (that's the dtype) where it's an int in the original data. This should then go away if warning when records are dropped because they're not part of a block.