neurodata/slbR: Statistical Learning Benchmarks

In assessing the generalizability of a statistical learning algorithm, it is vital to consider a variety of diverse, feature-rich datasets. In this package, we develop a simple interface to many common benchmark datasets, including the Penn Machine Learning Benchmarks Olson (2017) <arXiv:1703.00512>, UC-Irvine Machine-Learning Repository, and MNIST Lecun (1998) <doi:10.1109/5.726791> for both classification and regression tasks, allowing users to examine performance across many disparate contexts. Utilities are also included to automatically load and clean data from different datasets.

Getting started

Package details

MaintainerEric Bridgeford <[email protected]>
Package repositoryView on GitHub
Installation Install the latest version of this package by entering the following in R:
neurodata/slbR documentation built on May 22, 2019, 2:41 p.m.