The PamBinaries package is a set of functions for reading PAMGuard binary output (.pgdf files) into R. It is based on MATLAB functions created by Doug Gillespie and Michael Oswald, and is written to work in the same way.
Install the latest version from GitHub:
# make sure you have Rtools installed if(!require('devtools')) install.packages('devtools') # install from GitHub devtools::install_github('TaikiSan21/PamBinaries')
The main function of this package is
To read any PAMGuard binary file, just point this function to its location:
myBinaryFile <- './Files/SomeBinaryFile.pgdf' binaryData <- loadPamguardBinaryFile(myBinaryFile)
loadPamguardBinaryFile will automatically determine the type of data
stored in the binary file (Click Detector, DIFAR Processing, etc.) and call
the appropriate function.
binaryData is a list with two objects,
binaryData$data is a list with one element for every
individual item in the binary file, the contents of which will be different
depending on the type of binary file being read.
is a list containing some metadata and information used by the
There is also an option to reduce the size of objects read from a Click Detector, Whistle and Moan Detector, or DIFAR:
binaryDataSmall <- loadPamguardBinaryFile(myBinaryFile, skipLarge=TRUE)
This option will skip reading the waveform for click and difar data, and skip the contour and slice information for whistle data. This greatly reduces the size of the loaded object and the time required to read the data, useful if you need to process a large number of files.
You can also choose to only read in data with specific UIDs:
binarySomeUIDs <- loadPamguardBinaryFile(myBinaryFile, keepUIDs=c(100000001, 100000002))
This only read in the data with the UIDs provided in the keepUIDs argument, skipping over the rest. This can greatly reduce the time required to read data if you only need a small subset of the items in the binary file.
As of version 1.2 there is a second exported function,
pbToDf. This function converts
the output of
loadPamguardBinaryFile into a data frame. In order to do this it must
skip some of the information in the binaries that are not an appropriate size to put in
a data frame, as of version 1.2 the skipped data includes: any annotations data, click
wave forms, DIFAR demux data, and contour information from the Whistle & Moan Detector.
This function can be called directly, and is also called when as.data.frame is called
on an object of class
PamBinary; the output of
loadPamguardBinaryFile is both
PamBinary as of version 1.2.
The following should return identical results:
df1 <- pbToDf(binaryData) df2 <- as.data.frame(binaryData) df3 <- data.frame(binaryData) df4 <- pbToDf(binaryData$data)
The final example (
df4) is not the preferred way to make this conversion, but is
included to avoid possibly surprising behavior. The function will issue a warning,
but will convert to a data frame as expected.
PamBinaries should be compatible with Pamguard v2.00.15 and earlier.
- Added support for multiple matched classifier annotations
The output of
loadPamguardBinaryFileis now a
PamBinaryin addition to
list. This is just to allow for some easy method dispatch (like
as.data.framebelow), and should not change any other behavior.
Added a new exported function
pbToDfthat converts a
PamBinaryobject to a data frame.
Extended the generic
as.data.frame.PamBinaryis just a wrapper for
pbToDfbut allows for quick and easy conversion to data frames.
Slight change to function output, conversion of date to POSIXct object is no longer done automatically, only if the new flag
convertDateis set to
TRUE. Date is now reported as seconds since 1970-01-01 UTC, and there is a second exported function
convertPgDatethat will properly convert the numeric value to POSIXct. This change was made for speed purposes,
loadPamguardBinaryFilenow runs approximately 40% faster without the date conversion.
Added support for Click Trigger Background binary files.
- Initial release, see tutorial above.
Add rest of annotation readers. Will probably need to re-adjust existing ones in future PG versions.