| Title: | Estimate Human Activity from 'Accelerometry' Data |
| Version: | 0.2.0 |
| Description: | Interfaces the 'actinet' Python module https://github.com/OxWearables/actinet for an activity classification model based on self-supervised learning for wrist-worn accelerometer data. |
| License: | MIT + file LICENSE |
| Encoding: | UTF-8 |
| Imports: | assertthat, curl, magrittr, readr, reticulate (≥ 1.42.0) |
| Suggests: | tidyr, dplyr, ggplot2, testthat (≥ 3.0.0) |
| Config/testthat/edition: | 3 |
| Config/roxygen2/version: | 8.0.0 |
| Language: | en-US |
| URL: | https://github.com/jhuwit/actinet |
| BugReports: | https://github.com/jhuwit/actinet/issues |
| NeedsCompilation: | no |
| Packaged: | 2026-07-28 01:03:00 UTC; johnmuschelli |
| Author: | John Muschelli 
[aut, cre] |
| Maintainer: | John Muschelli <muschellij2@gmail.com> |
| Repository: | CRAN |
| Date/Publication: | 2026-08-05 17:50:02 UTC |
Load Actinet Model
Description
Load Actinet Model
Usage
ac_load_model(
classifier = c("walmsley", "willetts"),
model_path = NULL,
check_md5 = TRUE,
force_download = FALSE,
as_python = TRUE
)
ac_model_filename(classifier = c("walmsley", "willetts"))
ac_download_model(
model_path,
classifier = c("walmsley", "willetts"),
check_md5 = TRUE,
...
)
Arguments
classifier |
type of the model: either walmsley or willetts |
model_path |
the file path to the model. If on disk, this can be
re-used and not re-downloaded. If |
check_md5 |
Do a MD5 checksum on the file |
force_download |
force a download of the model, even if the file exists |
as_python |
Keep model object as a python object |
... |
for |
Value
A model from Python. ac_download_model returns a model file path.
Run Actinet Model on Data
Description
Run Actinet Model on Data
Usage
actinet(
file,
outdir = tempfile(),
classifier = NULL,
sample_rate = NULL,
model_path = NULL,
pytorch_device = NULL,
no_hmm = FALSE,
require_sleep_above = NULL,
single_sleep_block = FALSE,
force_download = FALSE,
exclude_first_last = NULL,
exclude_wear_below = NULL,
csv_start_row = NULL,
csv_txyz = NULL,
csv_txyz_idxs = NULL,
csv_date_format = NULL,
calibration_stdtol_min = NULL,
plot_activity = FALSE,
cache_classifier = FALSE,
verbose = TRUE
)
Arguments
file |
accelerometry file to process, including CSV,
CWA, GT3X, and |
outdir |
folder location to save output files |
classifier |
Enter custom activity classifier file to use. Default: walmsley (Walmsley2020 annotations of activity intensity). Can also enter path to local classifier (.joblib.lzma) file. |
sample_rate |
Sample rate for measurement, otherwise inferred. |
model_path |
the file path to the model. If on disk, this can be
re-used and not re-downloaded. If |
pytorch_device |
torch device to use, e.g.: 'cpu' or 'cuda:0'. Default: 'mps' if available, otherwise 'cpu' |
no_hmm |
Disable HMM post-processing |
require_sleep_above |
Require sleep blocks to exceed a minimum duration, otherwise be classified as sedentary. Pass values as strings, e.g.: '2H', '30min'. Default: None (no requirement) |
single_sleep_block |
Recognize only one sleep block per day, all other sleep blocks will be converted to sedentary |
force_download |
Force download of classifier file |
exclude_first_last |
first,last,both Exclude first, last or both days of data. Default: None (no exclusion) |
exclude_wear_below |
Exclude days with wear time below threshold. Pass values as strings, e.g.: '12H', '30min'. Default: None (no exclusion) |
csv_start_row |
Row number to start reading a CSV file. Default: 1 (First row) |
csv_txyz |
CSV_TXYZ Column names for time, x, y, z in CSV files. Comma_ separated string. Default: 'time,x,y,z' |
csv_txyz_idxs |
Column indices for time,x,y,z (0_indexed, e.g., '0,1,2,3'). Overrides csv_txyz. |
csv_date_format |
Date time format for csv file when reading a csv file. See https://docs.python.org/3/library/datetime.html#strftime_and_strptime_format_codes for more possible codes. Default: '%Y-%m-%d %H:%M:%S.%f' (e.g. '2023-10-01 12:34:56.789') |
calibration_stdtol_min |
Minimum standard deviation tolerance (g) for detecting stationary periods for calibration. Default: None |
plot_activity |
Plot the predicted activity labels |
cache_classifier |
Download and cache classifier file and model modules for offline usage |
verbose |
print diagnostic messages |
Value
A list of the results (data.frame),
summary of the results, adjusted summary of the results, and
information about the data.
Examples
library(magrittr)
file = system.file("extdata/P30_wrist100.csv.gz", package = "actinet")
if (actinet_check()) {
out = actinet(file = file)
data = readr::read_csv(out$outfiles[1])
daily_data = readr::read_csv(out$outfiles[3])
}
Check the actinet Python Module
Description
Check the actinet Python Module
Usage
have_actinet()
actinet_check()
actinet_version()
Value
A logical value indicating whether the actinet Python module is available.
Examples
if (have_actinet()) {
actinet_version()
}