one-class image classification implementation for julia

• this code is divided in two modules that can be used independently and one that orchestrates all the work:

  datafolder.jl

  • it loads a dataset in the form of

    folder (dataset)

    subfolder (class 0)

    subfolder (class 1)

  monovision.jl

  • it stores the architectures of the models available

  oculus.jl

  • it enables the training and inference of models

once you downloaded julia and have it opened on a terminal:

1. access to pkg typing "]", it will appear something like this:

(oculus.jl) pkg>

2. then you can activate the project with all the dependencies using activate .

et voilà