Download and import MNIST handwritten numeral datasets

get.mnist.R

get.mnist <- function(dir=NULL) {
  # dir: the path containing the extracted files:
  #  train-images-idx3-ubyte
  #  train-labels-idx1-ubyte
  #  t10k-images-idx3-ubyte
  #  t10k-labels-idx1-ubyte
  if(is.null(dir)) {
    require(R.utils)
    dir <- tempdir()
    u <- c('http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz', 
           'http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz',
           'http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz',
           'http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz')
    sapply(u, function(x) {
      download.file(x, f <- file.path(dir, basename(x)))
      gunzip(f)
    })
  }

  read <- function(type) {
    img <- file(file.path(dir, sprintf('%s-images-idx3-ubyte', type)), 'rb')
    hdr <- readBin(img, 'integer', n=4, endian="big")
    X <- t(sapply(seq_len(hdr[2]), function(ii) {
      readBin(img, 'integer', size=1, n=prod(hdr[3:4]), endian="big")
    }))  
    close(img)
    lab <- file(file.path(dir, sprintf('%s-labels-idx1-ubyte', type)), 'rb')
    readBin(lab, "integer", n = 1, size = 4, endian = "big")
    n <- readBin(lab, integer(), n=1, size=4, endian="big")
    y <- readBin(lab, integer(), n=n, size=1, signed=FALSE)  
    close(lab)
    cbind.data.frame(y=factor(y), X)
  }  
  train <- read('train')
  test <- read('t10k')
  list(train=train, test=test)
}