DataSet type needs # of unique keys and total # of keys

SLU-TMI / TextMining.jl

Other

24 stars 7 forks source link

Closed chucklesoclock closed 9 years ago

chucklesoclock commented 9 years ago

For Laplace smoothing

chucklesoclock commented 9 years ago

I think this is good with the mdata, right?

mtabor150 commented 9 years ago

I'm not so sure. It may make sense to just add a unique_keys field.

mtabor150 commented 9 years ago

In #65 I added features(# keys), and total(# total keys) in the Distribution type to take care of this.