DESCRIPTION

Package: imputation
Type: Package
Title: Missing Value imputation via weighted k-nearest neighbors (w-kNN)
Version: 0.7.6
Date: 2017-04-24
Authors@R: c(
  person("Alex", "Whitworth", email = "whitworth.alex@gmail.com",
    role = c("aut","cre")),
  person("Marcel", "Prince", email= "mprince@dandb.com", role= "ctb")
  )
URL: https://github.com/alexwhitworth/imputation
Description: Impute missing values in a data matrix using weighted-kNN.
    Weights in w-kNN are specified using a Gaussian kernel, specified by kernlab.
    Imputation may be done via parallel computing (v0.4). Canopies (ie-subsets) were
    implemented in (v0.6) and provide approximate solutions for large datasets...
    I suggest not using canopies on datasets with < ~10^5 observations and using
    canopies on datasets with > ~10^5 observations.
License: GPL
LazyLoad: yes
Depends:
    R (>= 3.2.1),
    locfit,
    kernlab,
    parallel (>= 3.0)
Suggests:
    testthat
LinkingTo: Rcpp
Imports:
    Rcpp
RoxygenNote: 5.0.1