GMVP Optimization Bounded K Clustering about ca_gmvp HOT 1 CLOSED

Sorry for taking it so long to reply you back.

1. Import cudf
It is much easier to install sklearn and it is more of a standard in ML community to use sklearn. So I actually updated my codes to use sklearn instead of cudf. Thank you for your suggestion.

2. GMVP_between_clusters function
It must be due to I used quotation mark ("", '') while using the argparse. For example it should have been --dim_reduction_method PCA, rather than --dim_reduction_method 'PCA'. The README.md file has been updated as well, so please check it out.

3. Components
The default values of no_of_PCA_components and no_of_tsne_components have been updated to 3, just for visualization purpose. The 3-long vectors can be easily visualized in the 3D plots. I would recommend you to experiment with the numbers. In my opinion, I do not think 252-long vectors are long, so we do not need to use dimensionality reduction methods such as PCA.

from ca_gmvp.