These are chat archives for thunder-project/thunder

7th
Sep 2016
Michael Churchill
@rmchurch
Sep 07 2016 14:43
@d-v-b I see, it looks like the version I was using (v0.4.0) was already using the mllib implementation. So I can do the same Series clustering using:
model = mllib.KMeans.train(data.rdd.values(), k=self.k, maxIterations=self.maxIterations)