Closed orhankislal closed 7 years ago
@iyerr3 The output class is the actual kmeans from the stats package, which means we are using the exactly same functions. Transition between R kmeans and madlib.kmeans should be quite easy. I can create a separate output class with its own functions but I am not sure if that will improve the usability.
@orhankislal Got it.
Now help me understand why we even need the setClass("kmeans.madlib")
if
we're not using that class.
On Mon, Sep 26, 2016 at 2:33 PM, Orhan Kislal notifications@github.com wrote:
@iyerr3 https://github.com/iyerr3 The output class is the actual kmeans from the stats package, which means we are using the exactly same functions. Transition between R kmeans and madlib.kmeans should be quite easy. I can create a separate output class with its own functions but I am not sure if that will improve the usability.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/pivotalsoftware/PivotalR/pull/46#issuecomment-249704260, or mute the thread https://github.com/notifications/unsubscribe-auth/ACIkB8YFKh1dpz3vWP5DGZ6NOcW_8kp5ks5quDoZgaJpZM4KAwON .
@iyerr3 Now I see the source of confusion. It is not necessary, just a leftover line from an earlier stage of development. I'll push a commit to remove it.
Please note that this wrapper requires the following commit for the MADlib to work: https://github.com/apache/incubator-madlib/pull/68/commits/d3e87b514112b324a9475b60f50a590e651f611c