Leland McInnes

You need to get a mapping from cluster labels as output to ids in the condensed tree. From there is is just a matter of following the tree (finding a...

Thanks for the updates. This looks like it is starting to come together. The test failures are due to importing networkx in the test module; but it doesn't seem to...

It is quite possible that these two features don't actually interact properly. I believe the approximate_predict is just using the clusters that would have been selected with the ``cluster_selection_epsilon`` argument,...

It may be a little tricky, but is probably feasible. I think for now the KNN classifier is more likely to the the useful approach. Sorry.

There isn't really anything built in to handle this unfortunately. The best thing to do would be to effectively build your own solution. You can certainly clustering a rolling period...

You may have issues if your ``min_cluster_size`` is large and your ``min_samples`` is not set. You could try setting ``min_samples`` to something smallish and see if that helps.

Haversine should be using Ball trees, and not require the full distance matrix. However, depending on the data it is possible that the ball tree search can still end up...

Yes, ``CondensedTree._select_clusters()`` selects out the clusters -- it returns the ids in the tree of the clusters that would get selected. One caveat: if you are using fancier selection approaches...

I believe it is somehow taking the dimension of your vectors to be 0, which is what would cause the problem. As to why it is doing that -- I...

As a side note, I would *really* recommend you do a dimension reduction step prior to running HDBSCAN here -- it won't do well with very high dimensional data (being...