6 pointsby marcandre2596 hours ago1 comment
  • marcandre2596 hours ago
    The project started as I wanted to learn more about the Expectation Maximisation (EM) algorithms. LCA's estimation algorithm is very similar to Gaussian Mixture Modeling, so it's a quite easy jump from popular online EM content.

    The project translates the functionalities of venerable polca R package. It's also nice to have polca as a reference package to validate output.

    I use Python at work, so the algorithm themselves are written in C++...