Applies the NeRV projection
NeRV(Data, lambda = 0.1, neighbors = 20, iterations = 10, cg_steps = 2, cg_steps_final = 40, randominit = T, OutputDimension = 2,
PlotIt = FALSE, Cls)
Matrix of the Data to be projected
Optional: Controls the trustworthiness-continuity tradeoff. Default = 0.1
Optional: Set the number of nearest neighbours that each point should have. Should be positive. Default = 20
Optional: The number of iterations to perform. Default = 10
Optional: The number of conjugate gradient steps to perform per iteration in NeRV's optimization scheme. Default = 2
Optional: The number of conjugate gradient steps to perform on the final iteration in NeRV's optimization scheme. Default = 40
Optional: TRUE: Random Initialization (default), FALSE: PCA initializiation
Optional: Number of dimensions in the Outputspace, default=2
Optional: Should the projected points be plotted? Default: FALSE. Note: this is only usefull if OutputDimension = 2.
Optional: Vector containing the number of the class for each row in Data. This is only used to color the points according to their classes if PlotIt = T
OutputDimension-dimensional matrix of projected points
Uses the NeRV projection with matrix Data and lambda. Lambda controls the trustworthiness-continuity tradeoff.
Jarkko Venna, Jaakko Peltonen, Kristian Nybo, Helena Aidos, and Samuel Kaski. Information Retrieval Perspective to Nonlinear Dimensionality Reduction for Data Visualization. Journal of Machine Learning Research, 11:451-490, 2010. Jarkko Venna and Samuel Kaski. Nonlinear Dimensionality Reduction as Information Retrieval. In Marina Meila and Xiaotong Shen, editors, Proceedings of AISTATS 2007, the 11th International Conference on Artificial Intelligence and Statistics. Omnipress, 2007. JMLR Workshop and Conference Proceedings, Volume 2: AISTATS 2007.