jimhester / largevis

An implementation of the largeVis algorithm for visualizing large, high-dimensional datasets, for R

largeVis

This is an implementation of the largeVis algorithm described in (https://arxiv.org/abs/1602.00370). It also incorporates:

• A very fast algorithm for estimating k-nearest neighbors, implemented in C++ with Rcpp and OpenMP. See the Benchmarks file for performance details.
• Efficient implementations of the clustering algorithms:
  • HDBSCAN
  • OPTICS
  • DBSCAN
• Functions for visualizing manifolds like this.

News Highlights

• Version 0.1.10 re-adds clustering, and also adds momentum training to largeVis, as well as a host of other features and improvements.
• Version 0.1.9.1 has been accepted by CRAN. Much grattitude to Uwe Ligges and Kurt Hornik for their assistance, advice, and patience.

Some Examples

Clustering With HDBSCAN

Visualize Embeddings

Building Notes

• The CRAN binaries are likely to have been compiled with 32-bit ARMA objects, and without OpenMP on OS X.

• To get 64-bit ARMA objects, add to ~.R/Makevars:

  R_XTRA_CXXFLAGS = -DARMA_64BIT_WORD
  

  and recompile from source. (devtools::install_github("elbamos/largeVis") will work).

• Getting OpenMP support on OS X is a bit tricky. What I've done, is to install llvm (version 3.8 or later), and then add the following to ~.R/Makevars:

  SHLIB_OPENMP_CFLAGS = -fopenmp
  R_XTRA_CXXFLAGS = -DARMA_64BIT_WORD
  LDFLAGS =  ""-L/usr/local/opt/llvm/lib -Wl,-rpath,/usr/local/opt/llvm/lib"
  CPPFLAGS =  -I/usr/local/opt/llvm/include
  PATH = /usr/local/opt/llvm/bin:$PATH 
  

• Recompile from source, as above.