ResNet.jl

ResNet implementation in Julia
Author pxl-th
Popularity
8 Stars
Updated Last
2 Years Ago
Started In
June 2020

ResNet

Create model:

model = ResNetModel(18; in_channels=3, classes=10)

or you can ommit classification layer, by specifying number of classes as nothing.

  • simple:
y = x |> model
  • extract list of features:
features = model(x, Val(:stages))

Number of channels for each element of features is in model.stages. Size of the model is in model.size.