luz

R-CMD-check Codecov test coverage Discord

luz is a higher level API for torch providing abstractions to allow for much less verbose training loops.

This package is in very early stage of development. Don’t use for anything meaningful yet.

It’s heavily inspired in other higher level frameworks for deep learning, to cite a few:

Installation

Luz is not yet available on CRAN. You can install the development version with:

remotes::install_github("mlverse/luz")

Example

Luz let’s you take your Torch nn_module definition and fit it to a dataloader, while handling the boring parts like moving data between devices, updating the weights, showing progress bars and tracking metrics.

Here’s an example defining and training an Autoencoder for the MNIST dataset. We selected parts of the code to highlight Luz functionality. You can find the full example code here.

net <- nn_module(
  "Net",
  initialize = function() {
    self$encoder <- nn_sequential(
      nn_conv2d(1, 6, kernel_size=5),
      nn_relu(),
      nn_conv2d(6, 16, kernel_size=5),
      nn_relu()
    )
    self$decoder <- nn_sequential(
      nn_conv_transpose2d(16, 6, kernel_size = 5),
      nn_relu(),
      nn_conv_transpose2d(6, 1, kernel_size = 5),
      nn_sigmoid()
    )
  },
  forward = function(x) {
    x %>%
      self$encoder() %>%
      self$decoder()
  }
)

Now that we have defined the Autoencder architecture using torch::nn_module(), we can fit it using Luz:

fitted <- net %>%
  setup(
    loss = nn_mse_loss(),
    optimizer = optim_adam
  ) %>%
  fit(train_dl, epochs = 1, valid_data = test_dl)