Skip to contents

Download MNIST

Source: R/mnist.R
download_mnist.Rd

Download MNIST database of handwritten digits.

Usage

download_mnist(
  base_url = mnist_url,
  verbose = FALSE,
  as = c("data.frame", "list"),
  timeout = 1800
)

Format

A data frame with 785 variables:

  • px1, px2, px3 ... px784: Integer pixel value, from 0 (white) to 255 (black).

  • Label: The digit represented by the image, in the range 0-9.

Pixels are organized row-wise. The Label variable is stored as a factor.

There are 70,000 digits in the data set. The first 60,000 are the training set, as found in the train-images-idx3-ubyte.gz file. The remaining 10,000 are the test set, from the t10k-images-idx3-ubyte.gz file.

Items in the dataset can be visualized with show_mnist_digit().

For more information about the original dataset see https://yann.lecun.com/exdb/mnist/.

Arguments

base_url

Base URL that the MNIST files are located at.

verbose

If TRUE, then download progress will be logged as a message.

as

Return format. Use "data.frame" for the original data frame shape, or "list" for the canonical image result.

timeout

Minimum download timeout in seconds. The default is 30 minutes; a larger existing global R timeout is preserved.

Value

If as = "data.frame", a data frame containing the MNIST digits. If as = "list", a canonical image result with an integer matrix in data and factor digit labels in meta$label.

Details

Downloads the image and label files for the training and test datasets from the https://github.com/fgnt/mnist mirror of the original MNIST files and converts them to a data frame or canonical list result.

Note

Originally based on a function by Brendan O'Connor.

Canonical list results

as = "list" returns a shallow list with data (one image per row), meta (one metadata row per image), image_dim, channel_order, and source. Metadata uses lower-case invariant names when applicable: label, description, split, id, object, and pose; dataset-specific fields are retained in meta. split is explicit, so train/test identity does not depend on row position. source records the dataset and acquisition URL.

Examples

if (FALSE) { # \dontrun{
# download the MNIST data set
mnist <- download_mnist()

# first 60,000 instances are the training set
mnist_train <- head(mnist, 60000)
# the remaining 10,000 are the test set
mnist_test <- tail(mnist, 10000)

# PCA on 1000 random training examples
mnist_r1000 <- mnist_train[sample(nrow(mnist_train), 1000), ]
pca <- prcomp(mnist_r1000[, 1:784], retx = TRUE, rank. = 2)
# plot the scores of the first two components
plot(pca$x[, 1:2], type = "n")
text(pca$x[, 1:2],
  labels = mnist_r1000$Label,
  col = rainbow(length(levels(mnist$Label)))[mnist_r1000$Label]
)
} # }