Support more of DRToolkit's functionality

[hackalog]

DR tooklit (Matlab) implements the following functions:

• Principal Component Analysis ('PCA')
• Linear Discriminant Analysis ('LDA')
• Multidimensional scaling ('MDS')
• Probabilistic PCA ('ProbPCA')
• Factor analysis ('FactorAnalysis')
• Sammon mapping ('Sammon')
• Isomap ('Isomap')
• Landmark Isomap ('LandmarkIsomap')
• Locally Linear Embedding ('LLE')
• Laplacian Eigenmaps ('Laplacian')
• Hessian LLE ('HessianLLE')
• Local Tangent Space Alignment ('LTSA')
• Diffusion maps ('DiffusionMaps')
• Kernel PCA ('KernelPCA')
• Generalized Discriminant Analysis ('KernelLDA')
• Stochastic Neighbor Embedding ('SNE')
• Symmetric Stochastic Neighbor Embedding ('SymSNE')
• t-Distributed Stochastic Neighbor Embedding ('tSNE')
• Neighborhood Preserving Embedding ('NPE')
• Locality Preserving Projection ('LPP')
• Stochastic Proximity Embedding ('SPE')
• Linear Local Tangent Space Alignment ('LLTSA')
• Conformal Eigenmaps ('CCA', implemented as an extension of LLE)
• Maximum Variance Unfolding ('MVU', implemented as an extension of LLE)
• Landmark Maximum Variance Unfolding ('LandmarkMVU')
• Fast Maximum Variance Unfolding ('FastMVU')
• Locally Linear Coordination ('LLC')
• Manifold charting ('ManifoldChart')
• Coordinated Factor Analysis ('CFA')
• Gaussian Process Latent Variable Model ('GPLVM')
• Deep autoencoders ('Autoencoder')
• Neighborhood Components Analysis ('NCA')
• Maximally Collapsing Metric Learning ('MCML')
• Large Margin Nearest Neighhbor metric learning ('LMNN')

It also contains 6 techniques for intrinsic dimensionality estimation:

• Eigenvalue-based estimation ('EigValue')
• Maximum Likelihood Estimator ('MLE')
• Estimator based on correlation dimension ('CorrDim')
• Estimator based on nearest neighbor evaluation ('NearNb')
• Estimator based on packing numbers ('PackingNumbers')
• Estimator based on geodesic minimum spanning tree ('GMST')

What, if any, should we support?