Skip to content
This repository has been archived by the owner on Jul 16, 2021. It is now read-only.

[WIP] Issue #152 use cholesky factorization #155

Open
wants to merge 22 commits into
base: master
Choose a base branch
from
Open

[WIP] Issue #152 use cholesky factorization #155

wants to merge 22 commits into from

Commits on Oct 17, 2016

  1. Use const for convergence epsilon

    @andrewcsmith
    andrewcsmith committed Oct 17, 2016
    Configuration menu
    Copy the full SHA
    9437fd3 View commit details
    Browse the repository at this point in the history

  2. Make new use with_weights

    @andrewcsmith
    andrewcsmith committed Oct 17, 2016
    Configuration menu
    Copy the full SHA
    67553dc View commit details
    Browse the repository at this point in the history

  3. bump deps

    @andrewcsmith
    andrewcsmith committed Oct 17, 2016
    Configuration menu
    Copy the full SHA
    3fe49dd View commit details
    Browse the repository at this point in the history

  4. Add an integration test for GMM

    @andrewcsmith
    andrewcsmith committed Oct 17, 2016
    Configuration menu
    Copy the full SHA
    92af698 View commit details
    Browse the repository at this point in the history

  5. Commit current status of GaussianMixtureModel rewrite

    @andrewcsmith
    andrewcsmith committed Oct 17, 2016
    Configuration menu
    Copy the full SHA
    3e564f8 View commit details
    Browse the repository at this point in the history

  6. Use mean of log_prob_norm for convergence

    @andrewcsmith
    andrewcsmith committed Oct 17, 2016
    Configuration menu
    Copy the full SHA
    d40b0bd View commit details
    Browse the repository at this point in the history

  7. Add cov regularization value to stabilize cholesky

    @andrewcsmith
    andrewcsmith committed Oct 17, 2016
    Configuration menu
    Copy the full SHA
    06ace3b View commit details
    Browse the repository at this point in the history

  8. Allow customization of initialization methods

    @andrewcsmith
    andrewcsmith committed Oct 17, 2016
    Configuration menu
    Copy the full SHA
    10bafa7 View commit details
    Browse the repository at this point in the history

  9. Investigate the initialized means

    @andrewcsmith
    andrewcsmith committed Oct 17, 2016
    Configuration menu
    Copy the full SHA
    7db6f0c View commit details
    Browse the repository at this point in the history

  10. Add logs, not multiply

    @andrewcsmith
    andrewcsmith committed Oct 17, 2016
    Configuration menu
    Copy the full SHA
    9fd6d36 View commit details
    Browse the repository at this point in the history

  11. Test for lower than epsilon rather than le(0)

    @andrewcsmith
    andrewcsmith committed Oct 17, 2016
    Configuration menu
    Copy the full SHA
    b3c136b View commit details
    Browse the repository at this point in the history

  12. Fix covariance calculation in update_gaussian_parameters

    @andrewcsmith
    andrewcsmith committed Oct 17, 2016
    Configuration menu
    Copy the full SHA
    f79c2f1 View commit details
    Browse the repository at this point in the history

  13. Fix lints, fix doctests

    @andrewcsmith
    andrewcsmith committed Oct 17, 2016
    Configuration menu
    Copy the full SHA
    918a740 View commit details
    Browse the repository at this point in the history

Commits on Oct 18, 2016

  1. Reorganize and add the cov regularization variable again

    @andrewcsmith
    andrewcsmith committed Oct 18, 2016
    Configuration menu
    Copy the full SHA
    9176560 View commit details
    Browse the repository at this point in the history

  2. Testing higher-dimensional data is more likely to fail, so we should … 

    …do it
    @andrewcsmith
    andrewcsmith committed Oct 18, 2016
    Configuration menu
    Copy the full SHA
    7adee14 View commit details
    Browse the repository at this point in the history

  3. Remove comments, lower coveriance

    @andrewcsmith
    andrewcsmith committed Oct 18, 2016
    Configuration menu
    Copy the full SHA
    2f20142 View commit details
    Browse the repository at this point in the history

  4. Test more clusters with GMM

    @andrewcsmith
    andrewcsmith committed Oct 18, 2016
    Configuration menu
    Copy the full SHA
    0b7f6d2 View commit details
    Browse the repository at this point in the history

  5. Clean up train method

    @andrewcsmith
    andrewcsmith committed Oct 18, 2016
    Configuration menu
    Copy the full SHA
    31b20ea View commit details
    Browse the repository at this point in the history

  6. Remove vestigial method initialize_covariances

    @andrewcsmith
    andrewcsmith committed Oct 18, 2016
    Configuration menu
    Copy the full SHA
    e54cc04 View commit details
    Browse the repository at this point in the history

  7. Move CovOption into a trait/struct design

    @andrewcsmith
    andrewcsmith committed Oct 18, 2016
    Configuration menu
    Copy the full SHA
    566c3ed View commit details
    Browse the repository at this point in the history

Commits on Oct 19, 2016

  1. Adding basic gmm example for verification

    @AtheMathmo
    AtheMathmo committed Oct 19, 2016
    Configuration menu
    Copy the full SHA
    fe8fa5b View commit details
    Browse the repository at this point in the history

  2. Merge pull request #1 from AtheMathmo/gmm-stability 

    Adding basic gmm example for verification
    @andrewcsmith
    andrewcsmith authored Oct 19, 2016
    Configuration menu
    Copy the full SHA
    9b18f1a View commit details
    Browse the repository at this point in the history