mlr3torch: A Deep Learning Framework in R based on mlr3 and torch
This is an incremental contribution providing a user-friendly deep learning interface for R users within the mlr3 ecosystem, but it does not introduce new methods or achieve SOTA results.
The paper introduces mlr3torch, an R package that integrates deep learning into the mlr3 ecosystem, enabling easy definition, training, and evaluation of neural networks for tabular and tensor data. It supports predefined architectures, graph-based model definition, and integration with mlr3's resampling and benchmarking tools, demonstrated through hyperparameter tuning, fine-tuning, and multimodal data use cases.
Deep learning (DL) has become a cornerstone of modern machine learning (ML) praxis. We introduce the R package mlr3torch, which is an extensible DL framework for the mlr3 ecosystem. It is built upon the torch package, and simplifies the definition, training, and evaluation of neural networks for both tabular data and generic tensors (e.g., images) for classification and regression. The package implements predefined architectures, and torch models can easily be converted to mlr3 learners. It also allows users to define neural networks as graphs. This representation is based on the graph language defined in mlr3pipelines and allows users to define the entire modeling workflow, including preprocessing, data augmentation, and network architecture, in a single graph. Through its integration into the mlr3 ecosystem, the package allows for convenient resampling, benchmarking, preprocessing, and more. We explain the package's design and features and show how to customize and extend it to new problems. Furthermore, we demonstrate the package's capabilities using three use cases, namely hyperparameter tuning, fine-tuning, and defining architectures for multimodal data. Finally, we present some runtime benchmarks.