Andrei Novikov

@kno10 , thanks, your performance test is different and it results look much better. By default pyclustering always uses C++ code. I have run it my local machine and got...

@kno10 , thanks for these results, I think it is much better to optimize python code as well, the problem that I wasn't use numpy for processing in case of...

Hi @Thalaivar , Yes, it is known issue, random sampling exists, but it not used by CURE algorithm, there is the same complain: #522 . There are plans to support...

@tongliuTL , the answer is quite simple, because I am the only developer in that open-source project and I am working on it during free-time only.

Hello @PascalEconomouCoupa , Yes, that's right, ROCK is adapted for numerical data only in pyclustering library. It is possible to support categorical data, but I do not have enough time...

Hello, @TyrandeWhisperwind , Euclidean distance is considered as a similarity between objects. What kind of similarity do you want to see? Currently, there is no way to specify another distance...

@TyrandeWhisperwind, I will introduce opportunity to use specific metric for CLARANS algorithm in the next release - 0.9.1. Thank you for the issue! - [ ] Argument to specify distance...

Depends on #643 .

Workaround implemented: user-defined metrics are forbidden for passing to CCORE, in case of user-defined metric python code is used.