megancatonferguson
commented
7 years ago Thank you! I'll give it a try and keep you posted.
On Wed, May 10, 2017 at 2:18 PM, DL Miller notifications@github.com wrote:
Fixed via f630aee.
Jason: please do let me know if that doesn't solve your issue!
On 08/05/2017 23:15, megancatonferguson wrote:
I am using R version 3.3.2 with mrds version 2.1.17. I am having trouble using int.range as a matrix in order to specify w separately for each observation. I made sure that the "distance" value for each observation was between the corresponding values of left and width in the int.range matrix.
Here is a sample of the code I used:
|int.range <- matrix(rep(0, (nrow(Bm.x95)*2)), nrow=nrow(Bm.x95)) colnames(int.range) <- c("left", "width") idx <- which(Bm.x95$VisSOPkm == 0.5) int.range[idx,2] <- 1.25 idx <- which(Bm.x95$VisSOPkm == 1.5) int.range[idx,2] <- 2.25 idx <- which(Bm.x95$VisSOPkm == 2.5) int.range[idx,2] <- 3.25 idx <- which(Bm.x95$VisSOPkm == 4) int.range[idx,2] <- 5.25 idx <- which(Bm.x95$VisSOPkm == 7.5) int.range[idx,2] <- 10.25 idx <- which(Bm.x95$VisSOPkm == 20) int.range[idx,2] <- 20 Bmi.ddfdwdist.trnc5pct.hr.super.catsize <- ddf(dsmodel = ~mcds(key = "hr", formula = ~super.catsize), data = Bm.x95, method = "ds", meta.data=list("int.range"=int.range), control=list("optimx.maxit"=10000, "refit"=TRUE, "nrefits"=40, debug=TRUE)) |
Here is a copy of the error that I received:
Warning in process.data(data, meta.data, check = FALSE) :
no truncation distance specified; using largest observed distance
Error in detfct.fit.opt(ddfobj, optim.options, bounds, misc.options) :
No convergence.
In addition: There were 26 warnings (use warnings() to see them)
> warnings()
Warning messages:
1 https://github.com/DistanceDevelopment/mrds/issues/1: In
detfct(distance, ddfobj, select, index, width, standardize, ... :
longer object length is not a multiple of shorter object length
I tried increasing the limits in the control list, but that didn't seem to solve the problem.
I tried troubleshooting the problem myself, using getAnywhere() to work my way through all of the functions, but couldn't figure out what was causing the error. The error arises in ddf.ds at this command: lt <- detfct.fit(ddfobj, optim.options, bounds, misc.options)
In detfct.fit, the error arises at this command: lt <- detfct.fit.opt(ddfobj, optim.options, bounds, misc.options)
In flpt.lnl, the following command also seems relevant: p1 <- distpdf(x$distance[!x$binned], ddfobj = ddfobj, select = !x$binned, width = width, standardize = FALSE, point = misc.options$point, left = left)
In fx, here's where the warning about incompatible objects arises: return(detfct(distance, ddfobj, select, index, width, standardize, stdint)/(width - left))
I'm happy to share my data if that would help troubleshoot. Thanks!
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/DistanceDevelopment/mrds/issues/16, or mute the thread https://github.com/notifications/unsubscribe-auth/AAAHoSLFYw_ 5t8Zn4QbGfHAJhy8cF2mmks5r34XlgaJpZM4NUggZ.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/DistanceDevelopment/mrds/issues/16#issuecomment-300615647, or mute the thread https://github.com/notifications/unsubscribe-auth/AafxNbF3p0mQjmc3249wF8ES7Qq2tMjrks5r4inAgaJpZM4NUggZ .
-- Megan C. Ferguson Cetacean Assessment and Ecology Program Marine Mammal Laboratory Alaska Fisheries Science Center National Marine Fisheries Service NOAA 7600 Sand Point Way NE Seattle, WA 98115-6349 USA Tel: 1.206.526.6274 Fax: 1.206.526.6615 Megan.Ferguson@noaa.gov
dill
I am using R version 3.3.2 with mrds version 2.1.17. I am having trouble using int.range as a matrix in order to specify w separately for each observation. I made sure that the "distance" value for each observation was between the corresponding values of left and width in the int.range matrix.
Here is a sample of the code I used:
Here is a copy of the error that I received:
Warning in process.data(data, meta.data, check = FALSE) :
I tried increasing the limits in the control list, but that didn't seem to solve the problem.
I tried troubleshooting the problem myself, using getAnywhere() to work my way through all of the functions, but couldn't figure out what was causing the error. The error arises in ddf.ds at this command: lt <- detfct.fit(ddfobj, optim.options, bounds, misc.options)
In detfct.fit, the error arises at this command: lt <- detfct.fit.opt(ddfobj, optim.options, bounds, misc.options)
In flpt.lnl, the following command also seems relevant: p1 <- distpdf(x$distance[!x$binned], ddfobj = ddfobj, select = !x$binned, width = width, standardize = FALSE, point = misc.options$point, left = left)
In fx, here's where the warning about incompatible objects arises: return(detfct(distance, ddfobj, select, index, width, standardize, stdint)/(width - left))
I'm happy to share my data if that would help troubleshoot. Thanks!