g.getmeta: function to extract meta-data (features) from data in...
In PeteJWatson/ggircal: Raw Accelerometer Data Analysis

Description Usage Arguments Value Author(s) References Examples

Reads a accelerometer file in blocks, extracts various features and stores average feature value per short or long epoch. Acceleration and angle metrics are stored at short epoch length. The non-wear indication score, the clipping score, temperature (if available), light (if available), and Euclidean norm are stored at long epoch length. The function has been designed and thoroughly tested with accelerometer files from GENEA and GENEActiv. Further, the function should be able to cope with csv-format data procuded by GENEActiv and Actigraph

g.getmeta(datafile, desiredtz = c(), 
windowsizes = c(5, 900, 3600),daylimit = FALSE, 
offset = c(0,0,0), scale = c(1,1,1),
tempoffset = c(0,0,0),do.bfen = FALSE, do.enmo = TRUE,
do.lfenmo = FALSE, do.en = FALSE,
do.hfen = FALSE, do.hfenplus = FALSE, do.mad = FALSE,
do.anglex=FALSE,do.angley=FALSE,do.anglez=FALSE,
do.roll_med_acc_x=FALSE,do.roll_med_acc_y=FALSE,do.roll_med_acc_z=FALSE,
do.dev_roll_med_acc_x=FALSE,do.dev_roll_med_acc_y=FALSE,do.dev_roll_med_acc_z=FALSE,
do.enmoa = FALSE,lb = 0.2,hb = 15, n = 4,meantempcal=c(),chunksize=c(),
selectdaysfile=c(),dayborder=0,dynrange=c(),...)

`datafile`	name of accelerometer file
`desiredtz`	desired timezone: see also http://en.wikipedia.org/wiki/Zone.tab
`windowsizes`	Three values to indicate the lengths of the windows as in c(window1,window2,window3): window1 is the short epoch length in seconds and by default 5 this is the time window over which acceleration and angle metrics are calculated, window2 is the long epoch length in seconds for which non-wear and signal clipping are defined, default 900. However, window3 is the window length of data used for non-wear detection and by default 3600 seconds. So, when window3 is larger than window2 we use overlapping windows, while if window2 equals window3 non-wear periods are assessed by non-overlapping windows.
`daylimit`	number of days to limit (roughly), if set to FALSE no daylimit will be applied
`offset`	offset correction value per axis, usage: value = scale(value,center = -offset, scale = 1/scale)
`scale`	scaling correction value per axis, usage: value = scale(value,center = -offset, scale = 1/scale)
`tempoffset`	temperature offset correction value per axis, usage: value = scale(value,center = -offset, scale = 1/scale) + scale(temperature, center = rep(averagetemperate,3), scale = 1/tempoffset)
`do.bfen`	if TRUE, calculate metric BFEN with band-pass filter configuration set by `lb` and `hb`
`do.enmo`	if TRUE (default), calculate metric ENMO with negative values rounded to zero
`do.lfenmo`	if TRUE, calculate metric LFENMO with low-pass filter configuration set by `hb`
`do.en`	if TRUE, calculate metric EN
`do.hfen`	if TRUE, calculate metric HFEN with low-pass filter configuration set by `hb`
`do.hfenplus`	if TRUE, calculate metric HFENplus with band-pass filter configuration set by `lb` and `hb`
`do.mad`	if TRUE, calculate metric MAD (Mean Amplitude Deviarion)
`do.anglex`	if TRUE, calculate the angle of the x-axis relative to the horizontal plane (degrees) utilizing all three axes
`do.angley`	if TRUE, calculate the angle of the y-axis relative to the horizontal plane (degrees) utilizing all three axes
`do.anglez`	if TRUE, calculate the angle of the z-axis relative to the horizontal plane (degrees) utilizing all three axes
`do.enmoa`	if TRUE (default), calculate metric ENMOa which is equal to metric ENMO but with the absolute taken from the Euclidean norm minus one.
`do.roll_med_acc_x`	if TRUE, calculate rolling median for the x axis
`do.roll_med_acc_y`	if TRUE, calculate rolling median for the y axis
`do.roll_med_acc_z`	if TRUE, calculate rolling median for the z axis
`do.dev_roll_med_acc_x`	if TRUE, calculate deviations from rolling median for the x axis
`do.dev_roll_med_acc_y`	if TRUE, calculate deviations from rolling median for the y axis
`do.dev_roll_med_acc_z`	if TRUE, calculate deviations from rolling median for the z axis
`lb`	lower boundary of the frequency filter (in Hertz)
`hb`	upper boundary of the frequency filter (in Hertz)
`n`	order of the frequency filter
`meantempcal`	mean temperature corresponding to the data as used for autocalibration. If autocalibration is not done or if temperature was not available then leave blank (default)
`chunksize`	number between 0.2 and 1 to specificy the size of chunks to be loaded as a fraction of a 24 hour period, e.g. 0.5 equals 12 hour chunks. The default is 1 (24 hrs). For machines with less than 4 Gb of RAM memory a value below 1 is recommended.
`selectdaysfile`	see g.part1
`dayborder`	see g.part1
`dynrange`	see g.part1
`...`	Please ignore. Only used by the code internally when called from within g.part1 with selectdaysfile specific.

`metalong`	dataframe with long epoch meta-data: EN, non-wear score, clipping score, temperature
`metashort`	dataframe with short epoch meta-data: timestamp and metric
`tooshort`	indicator of whether file was too short for processing (TRUE or FALSE)
`corrupt`	indicator of whether file was considered corrupt (TRUE or FALSE)

Vincent T van Hees <vincentvanhees@gmail.com>

van Hees VT, Gorzelniak L, Dean Leon EC, Eder M, Pias M, et al. (2013) Separating Movement and Gravity Components in an Acceleration Signal and Implications for the Assessment of Human Daily Physical Activity. PLoS ONE 8(4): e61691. doi:10.1371/journal.pone.0061691
Aittasalo M, Vaha-Ypya H, Vasankari T, Husu P, Jussila AM, and Sievanen H. Mean amplitude deviation calculated from raw acceleration data: a novel method for classifying the intensity of adolescents physical activity irrespective of accelerometer brand. BMC Sports Science, Medicine and Rehabilitation (2015).

## Not run: 
datafile = "C:/myfolder/testfile.bin"

#Extract meta-data:
M = g.getmeta(datafile)

#Inspect first couple of rows of long epoch length meta data:
print(M$metalong[1:5,])

#Inspect first couple of rows of short epoch length meta data:
print(M$metalong[1:5,])

## End(Not run)