Week 4: Error in Running “Exceed Probability” Section

[Ching To Chung]

Hi. I have downloaded the code from week 4. All parts worked fine except the Exceedance probability section.

At line 74, exc1 <- sapply(mod.eco.reg$marginals.fitted.values,FUN = function(marg){1 – inla.pmarginal(q = 1, marginal = marg)}) returned an empty list. This caused the subsequent lines to fail.

I have spent a long time attempting to debug but I still don’t know where the problem is. I have not amended the code in any way and it is run as-is after the data directory is properly set. I am sure that all libraries, including INLA and its dependencies such as terra and sf are all properly installed.

I am using INLA build 24.06.27.

[Soe Htike]

I’m also stuck with an error message in Exceedence Probability.

> exc1 <- sapply(mod.eco.reg$marginals.fitted.values,FUN = function(marg){1 – inla.pmarginal(q = 1, marginal = marg)})
> summary(exc1)
Length Class Mode
0 list list
> exc1.cutoff <- c(0, 0.8, 0.9, 0.95, 1)
> cat.exc1<- cut(unlist(exc1),breaks=exc1.cutoff,include.lowest=TRUE)
Error in cut.default(unlist(exc1), breaks = exc1.cutoff, include.lowest = TRUE) :
‘x’ must be numeric

I can move on the the later steps, but because of this error, I cannot combine plots in the final step.

[Pyae Thu Tun]

Hello friends,
I also faced the same error by running the original code and it took me a long time to search for different options online.
Here comes the solution for me and you can try it too.

## showing error after running the code
> cat.exc1<- cut(unlist(exc1),breaks=exc1.cutoff,include.lowest=TRUE)
Error in cut.default(unlist(exc1), breaks = exc1.cutoff, include.lowest = TRUE) :
‘x’ must be numeric

## check the type and structure of exc1
> str(exc1_numeric)
Error: object ‘exc1_numeric’ not found
## it seems the variable was not successfully created

## check the content of exc1 after the ‘sapply()’
> str(exc1)
list()
## exc1 is empty

## check the data of previous object
> str(mod.eco.reg$marginals.fitted.values)
NULL
## it was not generated by the model

## solution is to add options at the last part of the code
## instead of
mod.eco.reg <- inla(formula.eco.reg, family=”poisson”,
data=data.suicides, E=E,
control.compute=list(dic=TRUE))

## use the following
mod.eco.reg <- inla(formula.eco.reg, family=”poisson”,
data = data.suicides, E = E,
control.compute = list(dic = TRUE, cpo = TRUE, waic = TRUE, return.marginals.predictor = TRUE))
## after editing, all the lines running smoothly for me and hope you can solve this too 🙂

[Ching To Chung]

Thank you so much! You’re a genius!

I have also just noticed that the code generates an incorrect map of the local Moran’s I. I am opening another thread for it. Do you have any ideas why?

[Pyae Thu Tun]

Oh, you’re right, Toby. I missed checking the output file and the saved JPG file shows the same as you . Currently, I am traveling and cannot run the code. Please do let us know when you have solved it. Thanks 👍

[Nichcha Subdee]

Thank you so much, Pyae! I followed your suggestion and it worked for me as well! ^O^

[Myat Htoo Linn]

Thanks all bros!!
You are life saviours.
I am also resolved.

[Author]

Posts