xGPR

Efficient Bayesian machine learning and discriminant analysis for sequences, graphs and tabular data.

• What is xGPR?

Quick start

• Installation
• How xGPR approximates kernels
• Quickstart tutorial

Examples

• Examples

Available kernels

• Available kernels

Advanced / in-depth tutorials

• Datasets & model constructor: in-depth
  • Dataset construction: details
  • Setting up regression models
  • Choosing the number of RFFs; a brief digression
  • Setting hyperparameters
  • Setting up a model for convolution
• Tuning hyperparameters: in-depth
  • General observations on hyperparameter tuning
  • A quick note on hyperparameter tuning bounds
  • Approximate vs exact NMLL
  • NMLL tuning methods
• Model fitting: in-depth
  • Preconditioning
  • Fitting
• Clustering and visualizing data
  • Generating random features

Miscellaneous

• Frequently asked questions
• About / Contact

Citations

If using xGPR in research intended for publication, please cite either:

Linear-Scaling Kernels for Protein Sequences and Small Molecules Outperform Deep Learning While Providing Uncertainty Quantitation and Improved Interpretability Jonathan Parkinson and Wei Wang Journal of Chemical Information and Modeling 2023 63 (15), 4589-4601 DOI: 10.1021/acs.jcim.3c00601

OR the preprint:

Jonathan Parkinson, & Wei Wang. (2023). Linear-Scaling Kernels for Protein Sequences and Small Molecules Outperform Deep Learning While Providing Uncertainty Quantitation and Improved Interpretability. https://arxiv.org/abs/2302.03294

• Search Page