grseb9s/implicitlyconstrainedlda

Code to generate the paper 'Implicitly Constrained Linear DIscriminant Analysis'

These are the resources used to generate Jesse H. Krijthe & Marco Loog's "Implicitly Constrained Linear Discriminant Analysis". See below for instructions on how to rerun the experiments, generate the figures used in the paper and/or generate the paper. The methods used have been extracted as a separate library, called RSSL. For these experiments, version 0.2 was used.

#Main Requirements (used to generate the original paper) R (3.0.2) RSSL (commit: 81458902dd6644caca4e1f8c9f20c3473e4fa3e4) pdflatex

Suggested: pdfcrop Helvetica.ttf and HelveticaBold.ttf files to include fonts in the PDF figures.

#How to reproduce this paper You can run the individual experiments using:

make experiment_learningcurvesgaussian
make experiment_crossvalidation
make experiment_learningcurves

Or run them all using (not recommended)

make experiments

To generate the figures and tables:

make figures

To prepare figures and tables for the manuscript

make preparation

To generate the manuscript:

make paper

Or, all at once (not recommended)

make all