mlr3torch: A Deep Learning Framework in R based on mlr3 and torch

TL;DR

mlr3torch is an R package that provides an extensible deep learning framework integrated with mlr3, enabling flexible neural network modeling, training, and evaluation for various data types.

Contribution

It introduces a novel R package that simplifies deep learning workflows within the mlr3 ecosystem, supporting graph-based network definitions and comprehensive pipeline integration.

Findings

Supports classification and regression for tabular and tensor data.

Enables defining neural networks as graphs within mlr3pipelines.

Demonstrates capabilities through hyperparameter tuning, fine-tuning, and multimodal architectures.

Abstract

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.