GGIR part 4 output error trouble shoot

[Jaslie Wu]

Dear all,

Recently, I come across a part 4 error, which made my GGIR shell function stop at part 4 output. Could you please kindly help me with this?

The error message is:

Part 4

Error in seq.int (0,to0 ~from, by) : wrong sign in ‘by’ argument

To provide more context of my issue, below are some steps I took to trouble shoot:

Confirmed my R version is up to date. 

Confirmed Part 2,3 outputs can be successfully generated.

I then tried to run the Part 4 codes line by line , it got stuck when bringing in the guider. 

Then the confusing thing is, when I run the same codes and same guider with the first 20 participants's data, all part 2,3,4,5 outputs can be generated successfully. But when I run the whole 220+ participants' data, the error occurs. 

My codes are attached below:

GGIR(mode = c(1:5),

     datadir = "",

     outputdir = "",

     studyname = "BaseLine",

     f0 = 1, f1 = 222,

     configfile = c(""),

     overwrite = TRUE,

     do.report = c(2,4,5),

     acc.metric = "ENMO",

     windowsize = c(5,900,3600),

     idloc = 2,

     dayborder = 0,

     iglevels = TRUE,

     sensor.location = "wrist",

     do.enmo = TRUE,

     do.cal = TRUE,

     do.anglez = TRUE,

     printsummary = TRUE,

     strategy = 1,

     ndaywindow = 7,

     includedaycrit = 16,

     includedaycrit.part5 = 2/3,

     includenightcrit = 16,

     minimum_MM_length.part5 = 23,

     do.imp = TRUE,

     excludefirstlast = FALSE,

     excludefirst.part5 = FALSE,

     part5_agg2_60seconds = TRUE,

     anglethreshold = 5,

     timethreshold = 5,

    HASIB.algo = 'vanHees2015',

    relyonguider = TRUE,

    sleeplogidnum = FALSE,

    SleepwindowType = 'TimeInBed',

    nnights = 7,

    cleaningcode = 0,

    loglocation = "",

    colid = 1,

    coln1 = 4,

    ignorenonwear = FALSE,

    sleepefficiency.metric = 1,

     mvpathreshold = 93.2,

     boutcriter = 0.8,

     mvpadur = c(1,5,10),

     boutcriter.in = 0.9,

     boutcriter.lig = 0.8,

     boutcriter.mvpa = 0.8,

     threshold.lig = 45.8,

     threshold.mod = 93.2,

     threshold.vig = 418.3,

     boutdur.mvpa = c(1,5,10),

     boutdur.in = c(10,20,30),

     boutdur.lig = c(1,5,10),

     frag.metrics = 'all',

     qwindow = c(0,24),

     epochvalues2csv = TRUE,

     save_ms5rawlevels = TRUE,

     save_ms5raw_without_invalid = FALSE,

     timewindow = c ('MM'),

     criterror = 1,

     visualreport = TRUE,

     do.visual = TRUE,

     outliers.only = TRUE,

     dofirstpage = TRUE)

    do.part2.pdf = TRUE)

Thank you for your time. Looking forward to hearing from you.

Kind Regards,

Jaslie

[Vincent van Hees]

Dear Jaslier,

You are specifying loglocation = "" which is not a valid path to a csv file. Please specify NULL or c() to indicate that no sleep log was used or specify the fill path including name of your sleep diary, e.g. "folder/sleeplog.csv".

If that does not address the issue:

1. Can you create a reproducible example with just one or two files? This should make it easier to investigate the problem.
2. Does the issue still occur in GGIR release 3.1-0? I see your message is from a month ago, maybe it has been resolved in the latest release.l
3. Can you run GGIR with its default settings without specifying all these arguments?  So, just datadir and outputdir in new output folder or with the old config.csv deleted.
4. If yes to question 3, can you iteratively add arguments to identify which of the arguments triggers the error?

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...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/RpackageGGIR/5474b4af-e47e-4095-8d16-1c2f2360c96fn%40googlegroups.com.