Ee Kin

I can add make a PR if that is fine with the HDBSCAN team

correct me if I am wrong, it seems like the merge height function https://github.com/scikit-learn-contrib/hdbscan/blob/47ef913970b7506789b46d53d37945f97c08fedf/hdbscan/_prediction_utils.pyx#L118-L183 might merge the height in clusters that does not belong to the same nearest neighbour tree...

i'll be happy to make a PR if you will be able to review it once it is done, should I do that?

I have tried without the memorydatalayer, and its still slow in initializing the net input with data and also slow in forward pass.

for step1(b) it feels more like it is using either one of a-->b or b

will there be updates on this? if not can I make a new PR that adds this class to keras-contrib? @gabrieldemarmiesse @stante , will be enabling DiscriminativeLearningRate in general but...

@gabrieldemarmiesse is there a reason why we shouldn't add this into keras directly?

what format is needed for the 'extra_data' parameter? Does this save the label for the vector and is able to output the label rather than the vector itself?

Hi @greglandrum when atom order changes the chiralty also changes, even when bond order is fixed. Any chance there is a script for handling both random atom order and random...

here's one @greglandrum ``` import random from rdkit import Chem # there are faster ways to do this, but this is quick and easy. :-) def is_cyclic_permutation(l1,l2): l1 = '...