/DrLIM

Implemention of Dimensionality reduction by learning an invariant mapping in CVPR2006

Primary LanguageJupyter Notebook

DrLIM

Implemention of Dimensionality reduction by learning an invariant mapping in CVPR2006

  • In the abstract of MOCO, The DrLIM was introduced, which utilizes the perspective of contrastive learning.

Loss Function.

  • left expression describe "similar" and right expression describe "dissimilar"
    image

Model Architecture

image

Dataset

  • MNIST
    • train: 3000 + a (Augmentation of -6, -3, and 3 degree rotation plus 6 pixel dropout.) of each class
    • test: 400 of each class
  • training set
    • ex) img1, img2, label. img1, img2 is random shuffle of "4" and "9". and label is write under the description.
  • Using class of "4" and "9"
    • label 0: same class image
    • label 1: different class image

Result.

result