Package: eimpute 0.2.4

Zhe Gao

eimpute: Efficiently Impute Large Scale Incomplete Matrix

Efficiently impute large scale matrix with missing values via its unbiased low-rank matrix approximation. Our main approach is Hard-Impute algorithm proposed in <https://www.jmlr.org/papers/v11/mazumder10a.html>, which achieves highly computational advantage by truncated singular-value decomposition.

Authors:Zhe Gao [aut, cre], Jin Zhu [aut], Junxian Zhu [aut], Xueqin Wang [aut], Yixuan Qiu [cph], Gael Guennebaud [cph, ctb], Jitse Niesen [cph, ctb], Ray Gardner [ctb]

eimpute_0.2.4.tar.gz
eimpute_0.2.4.tar.gz(r-4.5-noble)eimpute_0.2.4.tar.gz(r-4.4-noble)
eimpute_0.2.4.tgz(r-4.4-emscripten)eimpute_0.2.4.tgz(r-4.3-emscripten)
eimpute.pdf |eimpute.html✨
eimpute/json (API)
NEWS

# Install 'eimpute' in R:
install.packages('eimpute', repos = 'https://cloud.r-project.org')
Uses libs:
  • openblas– Optimized BLAS
  • c++– GNU Standard C++ Library v3

On CRAN:

Conda:

This package does not link to any Github/Gitlab/R-forge repository. No issue tracker or development information is available.

openblascpp

2.70 score 281 downloads 5 exports 2 dependencies

Last updated 8 months agofrom:e1a037a34a. Checks:3 OK. Indexed: no.

TargetResultLatest binary
Doc / VignettesOKMar 20 2025
R-4.5-linux-x86_64OKMar 20 2025
R-4.4-linux-x86_64OKMar 20 2025

Exports:biscalebiscale.controleimputeincomplete.generatorr.search

Dependencies:RcppRcppEigen

eimpute: Efficiently IMPUTE Large Scale Incomplete Matrix

Rendered fromeimpute.Rmdusingknitr::rmarkdownon Mar 20 2025.

Last update: 2022-05-02
Started: 2020-03-20

Citation

To cite package ‘eimpute’ in publications use:

Gao Z, Zhu J, Zhu J, Wang X (2024). eimpute: Efficiently Impute Large Scale Incomplete Matrix. R package version 0.2.4, https://CRAN.R-project.org/package=eimpute.

Corresponding BibTeX entry: 

  @Manual{,
    title = {eimpute: Efficiently Impute Large Scale Incomplete
      Matrix},
    author = {Zhe Gao and Jin Zhu and Junxian Zhu and Xueqin Wang},
    year = {2024},
    note = {R package version 0.2.4},
    url = {https://CRAN.R-project.org/package=eimpute},
  }

Readme and manuals

eimpute: Efficiently IMPUTE Large Scale Incomplete Matrix

Introdution

Matrix completion is a procedure for imputing the missing elements in matrices by using the information of observed elements. This procedure can be visualized as:

Matrix completion has attracted a lot of attention, it is widely applied in:

  • tabular data imputation: recover the missing elements in data table;
  • recommend system: estimate users' potantial preference for items pending purchased;
  • image inpainting: inpaint the missing elements in digit images.

Software

A computationally efficient R package, eimpute is developed for matrix completion.

Installation

Install the stable version from CRAN:

install.packages("eimpute")
Advantage

In eimpute, matrix completion problem is solved by iteratively performing low-rank approximation and data calibration, which enjoy two admirable advantages:

  • unbiased low-rank approximation for incomplete matrix
  • less time consumption via truncated SVD Moreover, eimpute also supports flexible data standardization.

Compare eimpute and softimpute in systhesis datasets $X_{m \times m}$ with $p$ proportion missing observations:

  • $m$ is chosen as 1000, 2000, 3000, 4000
  • $p$ is chosen as 0.1, 0.5, 0.9.

In high dimension case, als method in softimpute is a little faster than eimpute in low proportion of missing observations, as the proportion of missing observations increase, rsvd method in eimpute have a better performance than softimpute in time cost and test error. Compare with two method in *eimpute, rsvd method is better than tsvd in time cost.

References

  • Rahul Mazumder, Trevor Hastie and Rob Tibshirani (2010) Spectra Regularization Algorithms for Learning Large Incomplete Matrices, Journal of Machine Learning Research 11 (2010) 2287-2322

  • Nathan Halko, Per-Gunnar Martinsson, Joel A. Tropp (2011) Finding Structure with Randomness: Probabilistic Algorithms for Constructing Approximate Matrix Decompositions, SIAM Review Vol. 53, num. 2, pp. 217-288

Bug report

Send an email to Zhe Gao at gaozh8@mail.ustc.edu.cn