NEWS

---

agcounts 0.7.0 (2026-06-01)

• Fix problem with agread as it was not evaluating the tz argument
• Update parser "pygt3x" to work with idle sleep mode ouput from actigraph/pygt3x in v0.7.1

agcounts 0.6.9

• Added debugging capabilities to C++ code and additional arguments to control spherecrit, sdcriter, and minloadcrit within the gcalibrateC function. Thanks to John Muschelli for the contribution!

agcounts 0.6.8

• Fix issue #20 related to C++ compiling by adding the Rcpp plugin for C++ 11.
• Update C++ code for gcalibrateC to resolve issue when there are NaN values in sphere data (issue #35)
• Resolve issue #36 when calErrorEnd never gets below 0.01.
• Change minloadcrit from 72 to 168 as recommended in issue #34

agcounts 0.6.7

• Update agcalibrate to speed up function by first converting to a data.table before merging with the time stamps.
• Add parameter imputeTimeGaps to agcalibrate for users to decide if zeros are added back after calibration.

agcounts 0.6.6 (2024-02-29)

• Prepare agcounts for a CRAN submission
• Fix problem with GGIR::g.calibrate needing GGIR::g.inspectfile results.

agcounts 0.6.5

• Update README.md file.
• Update code to allow interaction with data that is read into a tibble.
• Minor updates to the Shiny app

agcounts 0.6.4

• Updated the agcounts license with KU Copyright and Creative Commons (CC BY-NC-SA 4.0).
• Changed the parsers to reflect the Python and R package names.
• Created more flexibility for adding a Shiny app theme.
• Extending the .last_complete_epoch function after discovering a file that did not find a complete epoch.

agcounts 0.6.3

• Improved test coverage including tests for the Shiny Server
• Removed the Shiny App from the inst folder to only make it a callable app using agShinyDeployApp

agcounts 0.6.2

• Update the Shiny rawDataModule UI and Server to filter by AM or PM to make plotting faster
• Improved test coverage from 19 to 56% by adding tests for the Shiny UI, gcalibrateC, get_counts, agread.
• Updated the test_actilife_counts.R file using a new data set that was uploaded for gcalibrateC testing.

agcounts 0.6.1

• Added start_time and stop_time attributes to data read by the pygt3x calibrated reader.
• Added lubridate::floor_date to round data_start in case data doesn't start on the specified epoch.
• Updated Shiny app documentation by moving all importFrom documentation to the agShinyDeployApp function.
• Deployed the Shiny app to shinyapps.io and contained it in the inst/agcounts folder.

agcounts 0.6.0

• Developed a Shiny app to allow users to visualize, explore, and compare activity counts.
• Added the .calculate_raw_metrics function to calculate ENMO and MAD for the Shiny App.
• Added the .read_agd function to extract metadata and data from the ActiGraph agd file for the Shiny app.
• Changed the use of the bitwise '&' with Boolean operands to '&&' in the gcalibrate.cpp file.
• Updated README.md with the new agread methods using the parser argument.

agcounts 0.5.0

• Updated pygt3x python reader (ActiGraph removed the CalibratedReader class).
• Added a C++ version of the GGIR g.calibrate function called agcalibrate to be used within a piping structure with calculate_counts.
• Removed the agread S3 methods and replaced it with a switch statement.
• Added options for the user to choose a calibrated or uncalibrated parser.
• Removed test_calibrated_readers.R file since agread.ggir and agread.pygt3x are no longer used.
• Added codecov to the agcounts package.

agcounts 0.4.0

• Added Github R-CMD-check workflow.
• Created a S3 generic function to extract and handle idle sleep time.
• Created a S3 generic function to read in the raw acceleration data using calibrated readers.
• Replaced a for loop in the .resample function with Rcpp to reduce processing time.
• Added a package hexagon sticker and updated the README with the calibrated readers.
• Added a unit test to test the calibrated readers.

agcounts 0.3.0

• Updated documentation for a new release
• Added Paul R. Hibbing as a package contributor.

agcounts 0.2.4

• Added a unit test to ensure the agcounts package and Actilife counts match
• Added a unit test to check if the output of the algorithm is as expected
• Added testthat (>= 3.0.0) to the description file.

agcounts 0.2.3

• Changed package namespace to include magrittr as an import
• Fixed the printing order of verbose by changing the magrittr's pipe to eager pipe
• Added dateTimeAs = "write.csv" to the data.table::fwrite function to correctly display time stamps when data are exported to csv
• Moved all of the internal functions in agcounts to the helpers.R file

agcounts 0.2.2

• Swapped in min in a place where pmin was incorrectly used

agcounts 0.2.1

• Fixed automatic sampling frequency detection to remove assumption about file having >= 1000 rows

agcounts 0.2.0

• Split calculate_counts out of get_counts to allow more flexible programming for end-users
• Moved the dot functions to internal
• Revised some of the logic for filling in missing timestamps during count extraction
• Revised some dealings with timezones to try making things more consistent
• Automated the detection of sampling frequency

agcounts 0.1.0

• Initial release