Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Hotgym predictor, anomaly tests #675

Open
wants to merge 14 commits into
base: master
Choose a base branch
from
Open

Hotgym predictor, anomaly tests #675

wants to merge 14 commits into from

Commits on Sep 18, 2019

  1. Configuration menu
    Copy the full SHA
    5585038 View commit details
    Browse the repository at this point in the history
  2. Hotgym: use TM.anomaly with different modes

    remove explicit AnomalyLikelihood
    breznak committed Sep 18, 2019
    Configuration menu
    Copy the full SHA
    d885653 View commit details
    Browse the repository at this point in the history
  3. Hotgym: use SDRClassifier, Predictor WIP

    but Predictor learn() crashes in Release, however does not in Debug
    (-> hard to investigate)
    breznak committed Sep 18, 2019
    Configuration menu
    Copy the full SHA
    432f98a View commit details
    Browse the repository at this point in the history

Commits on Sep 20, 2019

  1. Configuration menu
    Copy the full SHA
    9f015f1 View commit details
    Browse the repository at this point in the history
  2. Hotgym: debugged crash with Predictor

    crash was on bad_alloc (memory allocation failed).
    Our encoding to labels (realToCategory_()) had
    undeflow which resulted in a huge UInt -> tried to allocate
    extremely large vector -> failed.
    breznak committed Sep 20, 2019
    Configuration menu
    Copy the full SHA
    97023a8 View commit details
    Browse the repository at this point in the history
  3. Configuration menu
    Copy the full SHA
    ae192bd View commit details
    Browse the repository at this point in the history
  4. Hotgym: fewer steps in Predictor, for faster runtime

    the 4 different step-ahead predictions took a lot of time
    breznak committed Sep 20, 2019
    Configuration menu
    Copy the full SHA
    0987941 View commit details
    Browse the repository at this point in the history
  5. Hotgym: Predictor not learning, still zero

    add assert to catch whether we learn something.
    breznak committed Sep 20, 2019
    Configuration menu
    Copy the full SHA
    0e1cc14 View commit details
    Browse the repository at this point in the history
  6. Predictor: add reproducible test for large labels

    this crashes the Predictor. Likely a stack-overflow,
    breznak committed Sep 20, 2019
    Configuration menu
    Copy the full SHA
    6c7678e View commit details
    Browse the repository at this point in the history
  7. Predictor: use type StepsAheadT for steps_

    limit to unsingned short (from UInt)
    breznak committed Sep 20, 2019
    Configuration menu
    Copy the full SHA
    285490b View commit details
    Browse the repository at this point in the history
  8. Classifier: numCategories_ is size_t

    to avoid overflow for large labels
    breznak committed Sep 20, 2019
    Configuration menu
    Copy the full SHA
    06f9701 View commit details
    Browse the repository at this point in the history

Commits on Nov 6, 2019

  1. Configuration menu
    Copy the full SHA
    4bed29d View commit details
    Browse the repository at this point in the history
  2. fix bindings for setAnomalyMode

    needed to specify the argument for the signiture to be correct
    breznak committed Nov 6, 2019
    Configuration menu
    Copy the full SHA
    f80a765 View commit details
    Browse the repository at this point in the history

Commits on Nov 7, 2019

  1. progress

    breznak committed Nov 7, 2019
    Configuration menu
    Copy the full SHA
    865f7e0 View commit details
    Browse the repository at this point in the history