Integrate K-Means Clustering with CRRao

[sourish-cmi]

Integrate K-means clustering with CRRao
Integrate K-means clustering with CRRao from Clustering.jl package.

The Clustering.jl package is weird because it wants data to be supplied as d x n, where d is the dimension of the data, i.e., number of variables, and n is the number of samples. However, this is the opposite practice of the Stat community. In the Statistics community, it must be supplied as n x d. So we need to fix it.

The possible solution would look like

container = fit(DataFrame, KMeansClustering(),K::Int64,...)

If somebody does not want to use all variables in the DataFrame, then the solution would look like

container = fit(VarName, DataFrame, KMeansClustering(),K::Int64,...)

Warning: The dimension of data input in Clustering.jl is n x d

[a-keshav]

Hey, is this issue still open? and if yes, could you please assign it to me?

[sourish-cmi]

Hey, is this issue still open? and if yes, could you please assign it to me?

Sure why not - you can try it.

[a-keshav]

I have submitted a PR for this issue. In this implementation, the function returns a 'KmeansResult' object. One hurdle I see with the current implementation is that the attributes of the object returned are also of the form (d x n). Do you believe that instead of passing the object, the clustering results would be better passed as tuples?