search

aparker92 / error_analyzers

0 stars 3 forks source link

Problems Running Code (from Reproducibility Group) #13

Open kimberlyle opened 10 years ago

kimberlyle commented 10 years ago

Hi analyzers! I'm from the reproducibility group. We're running the code that you guys wrote in R (ErrorDiagram.R) but there's some sort of error when running the social.dat example. Have you guys been able to run the code?

Also, I noticed in the social.dat that it's referring to a file in Bonghyun's computer. It would probably be better to refer it to the source file location of social.txt

johnrisko commented 10 years ago

What does the error say?

kimberlyle commented 10 years ago

Error in createErrorDiagram(socal.dat, start, test.start, finish, etas.CI, : object 'period' not found

kimberlyle commented 10 years ago

in the function createErrorDiagram(), is the line: !n.test = sum(times<period) - n.training supposed to be n.test = sum(times<testing.period) - n.training?

johnrisko commented 10 years ago

We encountered this error many times as well: Look at this line:

nu.CI=sapply(CI.dist,function(x){mean(CI.list<x)}) nu.CI=sort(nu.CI,decreasing=TRUE)

Make sure decreasing is set to TRUE rather than T

kimberlyle commented 10 years ago

I'm still getting the same error. Have you guys pushed the most updated file onto your repository?

GalaxyNight-day commented 10 years ago

Hi Kimberklye! sorry for the late response, I just woke up .... anyway, try https://github.com/aparker92/error_analyzers/blob/master/ErrorDiagram.R. the problem you had is exactly what we had on Thursday, and we fixed.

GalaxyNight-day commented 10 years ago

so please use createErrorDiagram function in https://github.com/aparker92/error_analyzers/blob/master/ErrorDiagram.R. and this will fix your problem. Sorry for the confusion