Hierarchical clustering?

[dreiss-isb]

Any plans / interest to add hierarchical clustering to this package? Or is that more appropriate for a different package? In that case, this package should be renamed to Kmeans or some such.

[johnmyleswhite]

Yes, I'd like to have hierarchical clustering in this package. I'd love a pull request, but will otherwise get to it over the course of the summer.

[MatthewRalston]

Definitely would love to see this!

[lindahua]

There is a PR #32 on this, which is being under review.

[diegozea]

There is already a function for hierarchical clustering, but the API doesn't follow the style of this package. Is It possible to...

• ...make Hclust a sub-type of ClusteringResult?
• ...rename cutree to assignments (with h and k keyword arguments)?
• ...define counts with h and k keyword arguments?
• ...define nclusters with h keyword argument?

Also, the name hierarchical sounds more Julian than hclust.

If it's ok, I could make a PR after #59 (and maybe before #60) get merged.

What do you think?

Thanks!

Best

[diegozea]

Bump!

[kmsquire]

I think that would be fine. Do you feel comfortable reviewing #59 and #60? I think they could probably be merged if someone reviewed them and gave some feedback (even a "looks good").

[diegozea]

After viewing the changes and comparing results with R, #59 looks good ;)

[ararslan]

Now that we have HClust in here, I'll close this as having been addressed.

[diegozea]

@ararslan Was there a recent merged PR addressing the documentation and API of HClust to close this?

[ararslan]

@diegozea Heirarchical clustering was introduced in #32. There's still an outstanding PR (#60) to add documentation, but since this issue is asking for heirarchical clustering functionality and that's now available, it seems to me that the issue has been effectively addressed.

[diegozea]

@ararslan What do you think about this proporsal: #4 (comment) ?
I'm sorry. I haven't had time to work on it yet...

[ararslan]

What you propose in that comment sounds great to me!