Open nthaydt opened 5 years ago
Fixed the indexing associated with alpha2. However, I think rather than by individual and day it should just be by individual at most.
Yeah, I agree with that. I think making behavior vary by day is too much...
Sent from my iPhone
On Feb 28, 2019, at 12:16 PM, Daniel J. Hocking notifications@github.com<mailto:notifications@github.com> wrote:
Fixed the indexing associated with alpha2. However, I think rather than by individual and day it should just be by individual at most.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/Hocking-Lab/cocanal-turtles/issues/11#issuecomment-468357628, or mute the threadhttps://github.com/notifications/unsubscribe-auth/Abb2DG1ubBfjj_QdvrfrCfCEFaOtC_fYks5vSA7egaJpZM4bS74Y.
commit: 6273ccd30cca281d440027710296561c7e977daf path: Code/BehavioralModel.R
The error seems to occur around line 37 in the model -> logit(p0[i, j, k, t]) <- alpha0[k, t] + (alpha2*C[i,k]) + eta[i,k]
I'm having trouble inserting the Cik matrix into the logit function and keep getting the following error: Error in checkForRemoteErrors(lapply(cl, recvResult)) : 3 nodes produced errors; first error: RUNTIME ERROR: Invalid vector argument to ilogit
JAGS doesn't seem to accept the vector provided with the logit function. Creating my own logit function was mentioned as a possible fix.