Skip to content

benliu961/HMEI2023-bxliu

Repository files navigation

5 by 400:
    The results of 5 by 400 is very good. The predictions are accurate. (classification working)

2 by 200:
    similar results to the 5 by 400.

rand 5 by 400:
    sigma: 2.8643e6
    nu: 0.1147
    rho: 34977

    sigma: 1.4739e6
    nu: 0.37443
    rho: 6688

    sigma: 2.3407e6
    nu: 2.9989
    rho: 9053

    sigma: 3.5659e6
    nu: 1.012
    rho: 41797

    sigma: 1.8181e6
    nu: 1.5859
    rho: 3771

    results are also good when categorized

rand 10 by 200:
    sigma: 3.2119e6
    nu: 0.92067
    rho: 7541

    sigma: 1.1503e6
    nu: 0.47416
    rho: 37495

    sigma: 2.7595e6
    nu: 2.903
    rho: 4224

    sigma: 2.5776e6
    nu: 2.3118
    rho: 14321

    sigma: 1.2835e6
    nu: 1.0533
    rho: 8779

    sigma: 3.3465e6
    nu: 2.5929
    rho: 4980

    sigma: 3.14e6
    nu: 2.8828
    rho: 9225

    sigma: 3.9627e6
    nu: 2.3391
    rho: 25686

    sigma: 3.7049e6
    nu: 1.7893
    rho: 4265

    sigma: 3.8864e6
    nu: 1.7795
    rho: 10691


Observations from loss distributions:
    Prediciting rho is consistent and good. All losses for rho are very close to a normal distribution of some kind.
    sigma and nu aren't as good. Some of the losses for sigma and nu do not look normal.
    When the images are categorized, all three perform well, including in the random category.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages