GGIR version 1.5-1

[Vincent van Hees]

GGIR version 1.5-1 now online with a few important updates:

- A 'beta-version' functionality to read cwa-format files from Axivity has been added. Special thanks to Evgeny Mirkes from the University of Leicester for writing and sharing this code.
- Bug fixed in correct timestamp interpretation for GENEActiv data, which was accidentally introduced in August 2016. It has timestamps, but in the wrong timezone. This only affects GENEActiv data
- Other updates are listed here

For those interested in seeing the underlying code and contributing: I copied the GGIR github repo from the ucl github organisation where it has lived for the past year to a new neutral github organisation I recently created: https://github.com/wadpac.

For the latest overview of GGIR related publications see here. Please email me when you spot a new publication that is not on the list yet.

Thanks, Vincent

[Vincent van Hees]

I have uploaded GGIR version 1.5-3 with some bug fixes last night, introduced in version 1.5-1 I uploaded last week.

If you are working in Windows then please wait a couple of days as Windows binaries need to be compiled by CRAN before they become active.

Error messages that should be resolved now:

- Error in Math.factor(Xi) : ‘cumsum’ not meaningful for factors
- Error in Rcpp::sourceCpp("src/numUnpack.cpp") : file not found: 'V:/Axivity/R/src/numUnpack.cpp'

- Error in g.readaccfile(filename = datafile, blocksize = blocksize, blocknumber = i,  :  object 'P' not found".

- Error in as.matrix(deci$rawxyz) : object 'deci' not found => in relation to g.wavread

Additionally, I have made a start with building a package vignette to offers new users a speed course in using GGIR: https://cran.r-project.org/web/packages/GGIR/vignettes/GGIR.html.

I am aware that this needs further improvement, but if you have tips on what should be in there or want to contribute yourself, then please get in touch.

I also added some contributing guidelines at https://github.com/wadpac/GGIR

Thanks, and sorry for the inconvenience the recent bugs may have caused.

Vincent

[diego....@sickkids.ca]

Hi Vincent - thanks for posting the new version, but I am still getting an error - now when it gets to part 3.

This is the error I get, and it seems to be related to the package 'zoo' that is called in for the interpolations:

Error in approx(wix, unlist(object[wix]), xout = wrng) : 

  need at least two non-NA values to interpolate

Do you have any leads on what is going on or ho to fix it?

Thanks

Diego Bassani

[Vincent van Hees]

Hi Diego,

I will need more information, because I did not get this error when testing the package.

What accelerometer brand does your file come from?

What is the data format you are working with, e.g. bin / csv / cwa / wav?

Does this error occur for one specific file or for all your files?

Was part 1 and 2 run with the same version of GGIR or were these done with previous versions of GGIR? If not, have you tried processing it all again with the latest version?

Were you able to process the file(s) with previous versions of GGIR?

What operating system are you using?

Thanks, Vincent

--
You received this message because you are subscribed to the Google Groups "R package GGIR" group.
To unsubscribe from this group and stop receiving emails from it, send an email to RpackageGGIR+unsubscribe@googlegroups.com.
To post to this group, send email to Rpacka...@googlegroups.com.
Visit this group at https://groups.google.com/group/RpackageGGIR.
To view this discussion on the web, visit https://groups.google.com/d/msgid/RpackageGGIR/983c6690-bab0-46fe-a791-01065e152f37%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Diego Bassani]

Hi Vincent,  

I have had the same issue with multiple files. To answer all your questions:

What accelerometer brand does your file come from?

ActiGraph GT3X+ ActiLife v6.13.3 Firmware v1.5.0

What is the data format you are working with, e.g. bin / csv / cwa / wav?

CSV

Does this error occur for one specific file or for all your files?

All files

Was part 1 and 2 run with the same version of GGIR or were these done with previous versions of GGIR? If not, have you tried processing it all again with the latest version?

Both in the same version, and in a previous version - I tried both when I realized there was a problem with the previous version of GGIR (before you fixed the bugs)

Were you able to process the file(s) with previous versions of GGIR?

Yes, but this is the first time using part 3 and 4.

What operating system are you using?

macOS Sierra Version 10.12.4

Hope this helps, 

Diego

---

This e-mail may contain confidential, personal and/or health information(information which may be subject to legal restrictions on use, retention and/or disclosure) for the sole use of the intended recipient. Any review or distribution by anyone other than the person for whom it was originally intended is strictly prohibited. If you have received this e-mail in error, please contact the sender and delete all copies.

[diego....@sickkids.ca]

Hi Vincent - any clues on what is happening here? Any help will be greatly appreciated.

Thanks

Diego

[Vincent van Hees]

Diego,

Please use argument do.enmo == TRUE. Feel free to use any other metrics in parallel to that, but it seems I made GGIR in a way that it requires at least metric ENMO to be available.

I think part3, part4 and part5 all work like this. They explicitly search for metric ENMO in order to work, while part2 should work with any metric.

Thanks for pointing this out, maybe at some point in the future someone can try to get part5 to work for any metric.

I managed to process the file you shared with me, so I think this issue is fixed.

Cheers,

Vincent

[Diego Bassani]

Great! Thanks!

D