ggperio Issue

[charlykl]

Hallo everyone,

I am using ggetho to analyze my mouse activity data and so far it is wokring quite fine, except one little issue: when I want to calculate the periodogram with ggperio, it gives me this warning:

50: In optimize(function(par) fn(par, ...)/con$fnscale, lower = lower, ... :
NA/Inf replaced by maximum positive value

I cannot figure out what exactly is the problem and what this warning means. It gives me a periodorgam though.

I would be greatful for any kind of help,
Thank you,
Charly

[qgeissmann]

Hello @charlykl ,
thanks for reporting. Is there any chance you can provide minimal data and code to reproduce your issue? It is very hard to debug for us otherwise,
Cheers!

[charlykl]

I am sorry, I did not mean to close the conversation. I am completly new to this and I did it by accident.

So I got now a set of test data that should reproduce the problem:

meta_data_test.csv

Test_Data.csv

Let me know if you need anything else ant thanky ou for your help.

[charlykl]

ah and now i figured out how to reopen it :)

[qgeissmann]

Hi @charlykl ,
No problem, and welcome to github! right. it would help a lot if you also provide minimal code to reproduce your error.
In other words, strip your code to just make it produce the error, and nothing extra. On our side, we can then run it and see if we reproduce your error, then fix it :),
Q

[charlykl]

oh yeah that makes sense. I also realised that my problem is not actually with ggperiod but with the periodogram. I always get those warnings and I am not sure why.

Here you go:

require("ggplot2")
require("dplyr")
require("behavr")
require("ggetho")
require("zeitgebr")

Test_data<-as.data.table(read.csv("Test_Data.csv",na.strings ="N.A.",dec=".",sep=",",header=TRUE),key="id")
Test_metadata<-as.data.table(read.csv("meta_data_test.csv",na.strings ="N.A.",dec=".",sep=",",header=TRUE),key="id")
Test_behavr<-behavr(Test_data,Test_data)

per_dt <- periodogram(Visits, Test_behavr, FUN = ls_periodogram, resample_rate = 1)