{"Encoding":"UTF-8","Package":"bigstatsr","Type":"Package","Title":"Statistical Tools for Filebacked Big Matrices","Version":"0.6.2","Date":"2018-08-13","Authors@R":"c(\nperson(\"Florian\", \"Priv\",\nemail = \"florian.prive.21@gmail.com\",\nrole = c(\"aut\", \"cre\")),\nperson(\"Michael\", \"Blum\",\nemail = \"michael.blum@univ-grenoble-alpes.fr\",\nrole = \"ths\"),\nperson(\"Hugues\", \"Aschard\",\nemail = \"hugues.aschard@pasteur.fr\",\nrole = \"ths\"))","Description":"Easy-to-use, efficient, flexible and scalable statistical tools.\nPackage bigstatsr provides and uses Filebacked Big Matrices via memory-mapping.\nIt provides for instance matrix operations, Principal Component Analysis,\nsparse linear supervised models, utility functions and more\n.","License":"GPL-3","Language":"en-US","LazyData":"TRUE","ByteCompile":"TRUE","Depends":"R (>= 3.3.0)","Imports":"bigreadr (>= 0.1.3), cowplot, doParallel, foreach, ggplot2 (>=\n3.0.0), graphics, methods, parallel, Rcpp, RSpectra, stats,\nutils","LinkingTo":"BH, Rcpp, RcppArmadillo","Suggests":"biglasso, bigmemory (>= 4.5.33), covr, glmnet, grid,\nModelMetrics, RcppEigen, RhpcBLASctl, spelling (>= 1.2),\ntestthat","RoxygenNote":"6.1.0","URL":"https://privefl.github.io/bigstatsr","BugReports":"https://github.com/privefl/bigstatsr/issues","Collate":"'AUC.R' 'FBM-attach.R' 'crochet.R' 'FBM.R' 'FBM-code256.R'\n'FBM-copy.R' 'RcppExports.R' 'SVD.R' 'apply-parallelize.R'\n'biglasso.R' 'bigstatsr.R' 'colstats.R' 'counts.R'\n'crossprodSelf.R' 'mult.R' 'plot.R' 'predict.R' 'randomSVD.R'\n'read.R' 'scaling.R' 'tcrossprodSelf.R' 'transpose.R'\n'univLinReg.R' 'univLogReg.R' 'utils-assert.R' 'utils.R'\n'zzz.R'","NeedsCompilation":"yes","Packaged":"2018-08-16 18:14:51 UTC; privef","Author":"Florian Priv [aut, cre],\nMichael Blum [ths],\nHugues Aschard [ths]","Maintainer":"Florian Priv ","Repository":"CRAN","Date/Publication":"2018-08-17 13:10:05 UTC","repoType":"cran","tarballUrl":"ftp://cran.r-project.org/pub/R/src/contrib/bigstatsr_0.6.2.tar.gz","jsonAuthors":[{"name":"Florian Priv","email":"florian.prive.21@gmail.com","maintainer":true},{"name":"Michael Blum","email":"michael.blum@univ-grenoble-alpes.fr","maintainer":false},{"name":"Hugues Aschard","email":"hugues.aschard@pasteur.fr","maintainer":false}],"readme":"[![Travis-CI Build Status](https://travis-ci.org/privefl/bigstatsr.svg?branch=master)](https://travis-ci.org/privefl/bigstatsr)\n[![AppVeyor Build Status](https://ci.appveyor.com/api/projects/status/github/privefl/bigstatsr?branch=master&svg=true)](https://ci.appveyor.com/project/privefl/bigstatsr)\n[![Coverage Status](https://img.shields.io/codecov/c/github/privefl/bigstatsr/master.svg)](https://codecov.io/github/privefl/bigstatsr?branch=master)\n[![CRAN_Status_Badge](http://www.r-pkg.org/badges/version/bigstatsr)](https://cran.r-project.org/package=bigstatsr)\n[![DOI](https://zenodo.org/badge/doi/10.1093/bioinformatics/bty185.svg)](https://doi.org/10.1093/bioinformatics/bty185)\n\n\n# bigstatsr\n\nR package {bigstatsr} provides functions for fast statistical analysis of large-scale data encoded as matrices. The package can handle matrices that are too large to fit in memory thanks to memory-mapping to binary files on disk. This is very similar to the format `big.matrix` provided by [R package {bigmemory}](https://github.com/kaneplusplus/bigmemory), which is **no longer used** by this package (see [the corresponding vignette](https://privefl.github.io/bigstatsr/articles/bigstatsr-and-bigmemory.html)).\n\n

\n\n[**LIST OF FEATURES**](https://privefl.github.io/bigstatsr/reference/index.html)\n\n**Note that most of the algorithms of this package don't handle missing values.**\n\n\n## Installation\n\n```r\n# For the current development version\ndevtools::install_github(\"privefl/bigstatsr\")\n```\n\n## Small example\n\n```r\nlibrary(bigstatsr)\n\n# Create the data on disk\nX <- FBM(5e3, 10e3, backingfile = \"test\")$save()\n# If you open a new session you can do\nX <- big_attach(\"test.rds\")\n\n# Fill it by chunks with random values\nU <- matrix(0, nrow(X), 5); U[] <- rnorm(length(U))\nV <- matrix(0, ncol(X), 5); V[] <- rnorm(length(V))\nNCORES <- nb_cores()\n# X = U V^T + E\nbig_apply(X, a.FUN = function(X, ind, U, V) {\n X[, ind] <- tcrossprod(U, V[ind, ]) + rnorm(nrow(X) * length(ind))\n NULL ## you don't want to return anything here\n}, a.combine = 'c', ncores = NCORES, U = U, V = V)\n# Check some values\nX[1:5, 1:5]\n\n# Compute first 10 PCs\nobj.svd <- big_randomSVD(X, fun.scaling = big_scale(), \n k = 10, ncores = NCORES)\nplot(obj.svd)\n\n# Cleanup\nunlink(paste0(\"test\", c(\".bk\", \".rds\")))\n```\n\nLearn more with this \n[introduction to package {bigstatsr}](https://privefl.github.io/R-presentation/bigstatsr.html).\n\n## Input format\n\nAs inputs, package {bigstatsr} uses [Filebacked Big Matrices (FBM)](https://privefl.github.io/bigstatsr/reference/FBM-class.html).\n\nTo memory-map character text files, see [package {mmapcharr}](https://github.com/privefl/mmapcharr).\n\n## Bug report / Help\n\nPlease open an issue if you find a bug.\nIf you want help using {bigstatsr}, please post on Stack Overflow with the tag *bigstatsr* (not yet created). [How to make a great R reproducible example?](https://stackoverflow.com/q/5963269/6103040)\n\n## Use cases\n\n### Parallelisation\n\nPackage {bigstatsr} uses package {foreach} for its parallelization tasks. Learn more on parallelism with {foreach} with [this tuto](https://privefl.github.io/blog/a-guide-to-parallelism-in-r/).\n\n- [Permute matrix columns in parallel](https://stackoverflow.com/q/48832010/6103040)\n\n- [Parallelized search until found](https://stackoverflow.com/q/49056271/6103040)\n\n### Large datasets\n\n- [Computing the null space of a bigmatrix](https://stackoverflow.com/questions/46253537/computing-the-null-space-of-a-bigmatrix-in-r/) (works if one dimension is not too large)\n\n- [Rowwise matrix multiplication](https://stackoverflow.com/q/48879643/6103040)\n\n- [Operating with a big.matrix](https://stackoverflow.com/q/42111876/6103040)","jobInfo":{"package":"bigstatsr","version":"0.6.2","parsingStatus":"success","parserVersion":1,"parsedAt":"2019-03-03T19:30:08+0000"}}

Description

Usage

Arguments

Value

Examples