In patperry/r-frame: Flexible Data Frames

knitr::opts_chunk$set(
    collapse = TRUE,
    comment = "#>",
    fig.path = "README-"
)
options(width = 80)
set.seed(0)

frame

frame is an R package providing a dataset type analogous to a data.frame supporting two major extensions: (1) associate keys with one or more components to each row; (2) use matrix-like objects as columns, including nested datasets.

Installation

Stable version

frame is available on CRAN. To install the latest released version, run the following command in R:

### install.packages("frame") # not yet, actually

Development version

To install the latest development version, run the following:

devtools::install_github("patperry/r-frame")

Usage

Datasets

The dataset type is like a data.frame but it allows matrix-like columns, including sparse matrices and nested datasets.

# dataset with a sparse matrix column
(x <- dataset(age = c(35, 70, 12, 42),
              color = c("red", "blue", "black", "green"),
              set = Matrix::sparseMatrix(i = c(1, 1, 2, 3, 3, 4),
                                         j = c(3, 2, 1, 3, 2, 1),
                                         x = c(2.8, -1.3, 7.1, 0.1, -5.1, 3.8),
                                         dimnames = list(NULL, c("a", "b", "c")))))

# dataset with a dataset column
(y <- dataset(value = rnorm(4), nested = x))

Keys

Datasets can have multi-component keys that uniquely identify each row. You can index a dataset just like a data.frame, or you can use key values to extract particular rows.

# set multi-component keys
keys(x) <- keyset(major = c("x", "x", "y", "y"),
                  minor = c(1, 2, 1, 3))

# show the data keys and values
print(x)

# index with keys
x[dataset(major = c("y", "x"),
          minor = c(  3,   1)), ]

Citation

Cite frame with the following BibTeX entry:

@Manual{,
    title = {frame: Data with Context},
    author = {Patrick O. Perry},
    year = {2018},
    note = {R package version 0.0.0},
}

Contributing

The project maintainer welcomes contributions in the form of feature requests, bug reports, comments, unit tests, vignettes, or other code. If you'd like to contribute, either