Package {sleeper}

Title: Estimate Sleep Status from Accelerometry Data
Version: 0.5.0
Description: Wraps the classifier from the Sundararajan (2021) <doi:10.1038/s41598-020-79217-x> to estimate sleep using a random forest. Users must download the model files from Sundararajan (2020) <doi:10.5281/zenodo.3752645> in order to use this method.
License: GPL (≥ 3)
Encoding: UTF-8
Depends: R (≥ 4.1.0)
Imports: assertthat, curl, dplyr, methods, reticulate, rlang
Suggests: callr, readr, testthat (≥ 3.0.0)
Config/roxygen2/version: 8.0.0
Config/testthat/edition: 3
NeedsCompilation: no
Packaged: 2026-09-08 16:50:58 UTC; johnmuschelli
Author: John Muschelli ORCID iD [aut, cre]
Maintainer: John Muschelli <muschellij2@gmail.com>
Repository: CRAN
Date/Publication: 2026-09-17 09:10:02 UTC

Run estimate_sleep with Python

Description

Run estimate_sleep with Python

Usage

py_estimate_sleep(
  ...,
  pyenv_function = function() {
     sleeper::py_require_sleeper()
 },
  show = FALSE
)

Arguments

...

arguments to pass to estimate_sleep

pyenv_function

function that loads the forest Python package. By default, it uses py_require_sleeper to import the package. If this function has an args argument, the output of pyenv_function will be re-assigned to args.

show

Logical, whether to show the standard output on the screen while the child process is running, passed to callr::r()

Value

Output from estimate_sleep(), a tibble of times and classification of sleep.

Command for py_require for the full sleeper workflow

Description

Command for py_require for the full sleeper workflow

Usage

py_require_sleeper(...)

Arguments

...

arguments to pass to reticulate::py_require()

Value

A logical value indicating whether the package is available.

Compact Example Sleep Models

Description

Compact Example Sleep Models

Usage

sl_example_model(
  model_dir = file.path(tempdir(), "sleeper-example-models"),
  overwrite = FALSE
)

Arguments

model_dir

Directory in which to save the example models.

overwrite

Logical; overwrite files that are already present.

Value

The downloaded model paths, or NULL when a download fails.

Examples

models = sl_example_model()

Get Sleep Features

Description

Get Sleep Features

Usage

sl_features(data)

Arguments

data

A data.frame with columns of timestamp, x, y, z

Value

A tibble of times and classification of sleep.

Examples


file = system.file("extdata", "example_data.csv.gz", package = "sleeper")
if (requireNamespace("readr", quietly = TRUE)) {
  data = readr::read_csv(file, n_max = 3600L)
  feat = sl_features(data)
}

Check for Downloaded Sleep Models

Description

Check for Downloaded Sleep Models

Usage

sl_have_models(model_dir)

sl_download_models(
  model_dir,
  quiet = FALSE,
  ...,
  overwrite = FALSE,
  folds = 1:5
)

Arguments

model_dir

path to the folder with models from doi:10.5281/zenodo.3752645

quiet

argument passed to curl::curl_download

...

additional arguments to pass to curl::curl_download

overwrite

logical, if TRUE will overwrite existing files

folds

should all model folds be downloaded? If only some folds, specify that (used almost exclusively for testing).

Value

A logical indicating all folds are downloaded.

Estimate Sleep from Wrist-Worn Accelerometry

Description

Estimate Sleep from Wrist-Worn Accelerometry

Usage

sl_python_modules_installed(component = c("models", "features"))

estimate_sleep(data, epoch = 30L, model_dir, cores = 1L)

Arguments

component

Python dependency set to check. The default, "models", checks all modules needed for sleep classification; "features" checks only the feature-extraction modules.

data

A data.frame with columns of timestamp, x, y, z

epoch

Time in seconds for the time interval estimate

model_dir

path to the folder with models from doi:10.5281/zenodo.3752645

cores

Number of workers to use for model prediction. Any non-zero integer accepted by Python joblib is allowed; -1 uses all available workers. Defaults to one worker for CRAN compatibility.

Value

A tibble of times and classification of category.

Examples


  if (sl_python_modules_installed("models")) {
    models = sl_example_model()
    if (!is.null(models) && requireNamespace("readr", quietly = TRUE)) {
      file = system.file("extdata", "example_data.csv.gz", package = "sleeper")
      data = readr::read_csv(file, n_max = 3600L)
      # Two minutes is sufficient to demonstrate the workflow and avoids
      # applying the forest to the full 12-hour example recording.
      try({estimate_sleep(data, model_dir = dirname(models$binary[1]))})
    }
  }