1. 16 Sep, 2021 3 commits
  2. 15 Sep, 2021 1 commit
    • Daniel Wysocki's avatar
      Fixing broadcasting issue for some models · d78a89cd
      Daniel Wysocki authored
      Broadcasting was broken for the beta and truncnorm models when the
      parameters' dimensions exceed 1.  This occurs when `where=True`, as
      `x[True].shape == (1,)+x.shape`, whereas the desired behavior is to
      maintain the original shape.  To fix this, we simply broadcast the
      `where` array to `parameters.shape`.
      d78a89cd
  3. 07 Sep, 2021 1 commit
  4. 02 Sep, 2021 1 commit
  5. 31 Aug, 2021 1 commit
  6. 19 Jul, 2021 2 commits
  7. 08 Jul, 2021 1 commit
  8. 08 Jun, 2021 2 commits
  9. 15 Apr, 2021 2 commits
  10. 14 Apr, 2021 7 commits
  11. 11 Mar, 2021 1 commit
  12. 02 Mar, 2021 2 commits
  13. 23 Feb, 2021 1 commit
  14. 18 Feb, 2021 3 commits
    • Daniel Wysocki's avatar
      Correcting the last commit · aef854cf
      Daniel Wysocki authored
      aef854cf
    • Daniel Wysocki's avatar
      Fixing clash between --unique and --max-samples for copy routine · 9433434d
      Daniel Wysocki authored
      When combining `--unique` and `--max-samples`, the potentially reduced
      number of samples from limiting to unique samples causes the random
      selection to satisfy `--max-samples` to potentially attempt to index
      outside the array of remaining samples.
      9433434d
    • Daniel Wysocki's avatar
      Making initialization from cleaned samples faster · 57b53a89
      Daniel Wysocki authored
      Now we step through the samples in order, terminating once enough
      samples have posterior support to initialize all walkers.  This has
      the side effect of removing the shuffling that was previously done.
      If your input samples are not independent, be sure to shuffle them
      ahead of time, e.g., with `pop_models_h5_cleaned_samples post_cleaned.hdf5 copy`
      and `--max-samples` set less than the total.
      57b53a89
  15. 17 Feb, 2021 1 commit
  16. 16 Feb, 2021 4 commits
  17. 12 Feb, 2021 2 commits
  18. 15 Jan, 2021 2 commits
    • Daniel Wysocki's avatar
      Added an option to only keep unique samples in H5Clean copy script · 2fe686a8
      Daniel Wysocki authored
      The `copy` subcommand of `pop_models_h5_cleaned_samples` now has an
      optional `--unique` flag which discards any duplicate posterior
      samples.  This is useful when dealing with runs with stuck MCMC
      walkers, particularly when using such a run to seed another run in
      hopes of becoming un-stuck.
      2fe686a8
    • Daniel Wysocki's avatar
      Allowing use of h5py 3.0 and above · 6a12ea66
      Daniel Wysocki authored
      Previous dependency list did not allow h5py 3.0 or above as it was not
      released and could break compatibility.  Now that it is out and does
      not appear to break anything, released prior to the future 4.0 release
      are allowed.
      6a12ea66
  19. 17 Dec, 2020 1 commit
  20. 09 Dec, 2020 2 commits