klustakwik | UNMAINTAINED - use klusta | Machine Learning library

klustakwik is a C++ library typically used in Artificial Intelligence, Machine Learning, Numpy applications. klustakwik has no bugs, it has no vulnerabilities and it has low support. You can download it from GitHub.

KlustaKwik is a program for cluster analysis by fitting a mixture of Gaussians. It was designed for the specific problem of spike sorting of multi-electrode arrays, but can be used for any application. Technically, KlustaKwik works by implementing a hard EM algorithm with unconstrained covariance matrices. It also uses a number of tricks to greatly speed up execution. It has recently been updated to deal with high-dimensional data as required for spike sorting from large, dense electrode arrays. An older version of the code, which cannot deal with high-density arrays, can be found [here] But don’t use that version, use this one: even for plain tetrode data, the new version is fully backward compatible, and is also much faster. For spike sorting, KlustaKwik is designed to be used in conjunction with [SpikeDetekt] for spike detection, and [KlustaViewa] for manual verification and adjustment of clustering results. You may subscribe to our googlegroups discussions by sending an email to: klustaviewas+subscribe at googlegroups.com.