Machine Learning in Elixir

Identifying Cats and Dogs Again

To understand the powers of transfer learning, you’ll make a few adjustments to your model training code from the previous chapter and compare the performance of your newly trained model with the performance of the models you trained previously. To start, fire up a Livebook and install the following dependencies:

Mix.install([
  {:axon_onnx, "~> 0.4"},
  {:axon, "~> 0.5"},
  {:nx, "~> 0.5"},
  {:exla, "~> 0.5"},
  {:stb_image, "~> 0.6"},
  {:kino, "~> 0.8"},
])

You should recognize all of these dependencies from the previous chapters of this book with the exception of :axon_onnx, which is a library for importing and exporting Open Neural Network Exchange (ONNX) models to ...