There was a bug when running part 6. Argument f0 is larger than the number of files in meta/ms5.outraw.

[forget999]

Dear Developer, When using version 3.1-11, run mode = 1:6, no report is generated in the file summary report folder of the result file, although I set the part6_threshold_combi parameter. The console displays an error as shown below. My visualization settings are as follows,
dofirstpage =T,
do.visual = TRUE,
visualreport = TRUE,
old_visualreport = T,
part6DFA=T,
part6CR=T,
cosinor=T,
part6Window=c("W1", "W-1"),
part6_threshold_combi="40_100_120".

I'm not sure if this is a problem with version 3.1-11. I look forward to your reply. Thank you!

[jhmigueles]

You don't need to run part 6 unless you are specifically interested in the analyses occurring there, which I assume you are not from your original question.

Please, (1) reset mode = 1:5, remove the config file, and add only the parameter part6_threshold_combi to your original script and let me know if that works.

[forget999]

You don't need to run part 6 unless you are specifically interested in the analyses occurring there, which I assume you are not from your original question.

Please, (1) reset mode = 1:5, remove the config file, and add only the parameter part6_threshold_combi to your original script and let me know if that works.

Thank you for your prompt reply. I just followed your suggestion and re-ran the test. Visual reports were successfully generated. However, my current research focuses on the results of part 6. I would like to know how to run part 6 successfully?

[jhmigueles]

Please see chapters 13 and 14 in the GGIR digital book: https://wadpac.github.io/GGIR/

Also see specific documentation on parameters you are interested in the GGIR documentation help file:

library(GGIR)
?GGIR

[forget999]

Thanks again for your reply, I have solved the problem. Looking forward to more progress from GGIR!

[vincentvanhees]

@forget999 could you please share with us what solved the problem for you?

Doing so will help:

• Future GGIR users with the same problem reading this thread.
• Us to understand what aspect of GGIR functionality or documentation is not clear enough yet.

[forget999]

@forget999 could you please share with us what solved the problem for you?

Doing so will help:

• Future GGIR users with the same problem reading this thread.
• Us to understand what aspect of GGIR functionality or documentation is not clear enough yet.

Thanks for your reply! I have reset the "part6_threshold_combi" parameters number to meet the physical activity requirements of my research.
In addition, I think there is one thing that needs to be improved. The new version v3.1-11 of GGIR takes too long time to run part6, which takes an hour. But in the version v3.1-5, it is quite fast. I don't know where the difference is, or what adjustments can speed up the running speed of part6. If you have any suggestions, please let me know. Thank you!

[forget999]

@forget999 could you please share with us what solved the problem for you?

Doing so will help:

• Future GGIR users with the same problem reading this thread.
• Us to understand what aspect of GGIR functionality or documentation is not clear enough yet.

In addition, I have two additional suggestions for GGIR, which I hope will be considered in the next version.

1. In a special case, if the data record of a person's accelerometer is split, that is, there are more than 100 csv files in datadir. First, I need to merge these more than 100 csv files, and then run GGIR. Does GGIR consider the function of automatically merging multiple csv files in one file?
2. The Intradaily Variability (IV) and Interdaily Stability (IS) results ​​of part2 and part6 are different. Are they based on different algorithms? Although the values ​​are not much different, which one should I believe?

[vincentvanhees]

thanks

I have reset the "part6_threshold_combi" parameters number to meet the physical activity requirements of my research.

Yes, there was a bug in automating this, which has now been fixed by Jairo and I just merged it into the master branch. In the next GGIR release you can use part6_threshold_combi 's default setting of NULL again which will automatically use the first value of each intensity levels thresholds.

In addition, I think there is one thing that needs to be improved. The new version v3.1-11 of GGIR takes too long time to run part6, which takes an hour. But in the version v3.1-5, it is quite fast. I don't know where the difference is, or what adjustments can speed up the running speed of part6. If you have any suggestions, please let me know. Thank you!

Thanks for reporting, I have created a new issue for it as this was an unintentional side effect of making GGIR compatible to older R versions.

In a special case, if the data record of a person's accelerometer is split, that is, there are more than 100 csv files in datadir. First, I need to merge these more than 100 csv files, and then run GGIR. Does GGIR consider the function of automatically merging multiple csv files in one file?

See parameter maxRecordingInterval.

The Intradaily Variability (IV) and Interdaily Stability (IS) results ​​of part2 and part6 are different. Are they based on different algorithms? Although the values ​​are not much different, which one should I believe?

This difference is expected. The IS and IV algorithm is identical but applied to different segments of the recording, which I have tried to better explain in the latest revision of the corresponding documentation.

[forget999]

thanks

I have reset the "part6_threshold_combi" parameters number to meet the physical activity requirements of my research.

Yes, there was a bug in automating this, which has now been fixed by Jairo and I just merged it into the master branch. In the next GGIR release you can use part6_threshold_combi 's default setting of NULL again which will automatically use the first value of each intensity levels thresholds.

In addition, I think there is one thing that needs to be improved. The new version v3.1-11 of GGIR takes too long time to run part6, which takes an hour. But in the version v3.1-5, it is quite fast. I don't know where the difference is, or what adjustments can speed up the running speed of part6. If you have any suggestions, please let me know. Thank you!

Thanks for reporting, I have created a new issue for it as this was an unintentional side effect of making GGIR compatible to older R versions.

In a special case, if the data record of a person's accelerometer is split, that is, there are more than 100 csv files in datadir. First, I need to merge these more than 100 csv files, and then run GGIR. Does GGIR consider the function of automatically merging multiple csv files in one file?

See parameter maxRecordingInterval.

The Intradaily Variability (IV) and Interdaily Stability (IS) results ​​of part2 and part6 are different. Are they based on different algorithms? Although the values ​​are not much different, which one should I believe?

This difference is expected. The IS and IV algorithm is identical but applied to different segments of the recording, which I have tried to better explain in the latest revision of the corresponding documentation.

Thanks for your reply! I tried to set idloc and maxRecordingInterval to solve the problem of merging multiple measurement files (csv) under the same ID.
As shown in the figure below, these 5 csv files are measurements of the same person in consecutive time periods. I hope GGIR can merge them and run them. However, it failed. I found that idloc does not work. My settings are idloc=7 and maxRecordingInterval=0 to indicate that the extraction ID is "GT3XPLUS", suggesting that these files belong to the measurement data of the same person. However, "meta\basic" in the result file still shows the original names of independent these files, not the a single file with named "meta_GT3XPLUS".

At the same time, I am confused about the setting of maxRecordingInterval parameter. For example, for an uninterrupted continuous measurement of a person without missing time, the result csv is only divided into multiple parts. Should maxRecordingInterval be set to 0 at this time?