Right. Um well the tree was made by ape2ouch(archotree) (also in the
workspace file). I'm not sure what you mean by code...
I have indeed read ?hansen. All I could find was:
"Each matrix can be parameterized by nchar*(nchar+1) parameters, where
nchar is the number of quantitative characters."
I have one character, so as far as I could understand it that would
imply "2" should be specified. I tried that to no avail.
Cheers
On Mon, Jul 25, 2011 at 2:37 PM, Aaron A. King <kin...@umich.edu> wrote:
> I note that Protorosaurus_speneri sits at an unspecified time
>
> print(archotreeouch)
> as(archotreeouch,"data.frame")[137,]
>
> It would be possible for me to verify this is the only problem were you to
> provide code that reproduces your problem from scratch. In particular, since
> I cannot see how you constructed archotreeouch, I cannot verify that this is
> indeed the only problem.
>
> Regarding the initialization of the hansen optimizer (sqrt.alpha, sigma), I
> assume you have read '?hansen'. Can you be more specific as to what is not
> clear to you?
>
> On Monday, July 25, 2011 08:53:12 AM Roland Sookias wrote:
>> OK, well for brown:
>>
>> brown(logflatnodes,archotreeouch)
>>
>> and for hansen:
>>
>> hansen1=hansen(logflatnodes,archotreeouch,diet,sqrt.alpha=0,sigma=1)
>>
>> I also attach the workspace file. The table from which the diet and log fl
>> (femur length) data comes from is called "dietandflatnodes".
>>
>> Cheers
>>
>> On Mon, Jul 25, 2011 at 1:37 PM, Aaron King <kin...@umich.edu> wrote:
>> > Hi Roland,
>> >
>> > It looks like it probably has to do ith your data. If you send a code
>> > that reproduces the error, we can help you.
>> >
>> > Aaron
>> >
>> > On Jul 25, 2011, at 7:44, Roland Sookias <r.sook...@gmail.com> wrote:
>> > > Hi
>> > >
>> > > I posted this on r-help by accident, and realize it's more appropriate
>> >
>> > here,
>> >
>> > > so I'll send it here and do ignore the other post.
>> > >
>> > > I'm trying to use ouch's hansen and brown functions but I get the error:
>> > >> brown(logflatnodes,archotreeouch)
>> > >
>> > > Error in backsolve(l, x, k = k, upper.tri = upper.tri, transpose =
>> > > transpose) :
>> > > NA/NaN/Inf in foreign function call (arg 1)
>> > >
>> > > and with hansen also:
>> > >
>> > > Error in optim(par = c(sqrt.alpha, sigma), fn = function(par) { :
>> > > function cannot be evaluated at initial parameters
>> > >
>> > > I'm not sure what I'm supposed to put in the sqrt.alpha and sigma
>> > > parameters. I have several different selective regimes (trophic
>> > > guilds).
>> >
>> > I
>> >
>> > > have a single continuous character (body size). I read these data in as
>> > > a table, and then made the columns into objects (or vectors? i.e. I
>> > > did data2=allmydata$data2) and named them by node number. The tree is
>> >
>> > converted
>> >
>> > > from ape.
>> > >
>> > > Thanks very much, and happy to provide more info...
>> > >
>> > > Traceback for the first error gives:
>> > >> traceback()
>> > >
>> > > 9: .C("bakslv", t = r, ldt = nrow(r), n = k, b = x, ldb = k, nb = nb,
>> > >
>> > > x = matrix(0, k, nb), job = job, info = integer(1L), DUP = FALSE,
>> > > PACKAGE = "base")
>> > >
>> > > 8: backsolve(l, x, k = k, upper.tri = upper.tri, transpose = transpose)
>> > > 7: forwardsolve(vh, a, upper.tri = TRUE, transpose = TRUE)
>> > > 6: as.matrix(x)
>> > > 5: svd(forwardsolve(vh, a, upper.tri = TRUE, transpose = TRUE))
>> > > 4: glssoln(w, x, b)
>> > > 3: FUN(X[[1L]], ...)
>> > > 2: lapply(dat, function(x) glssoln(w, x, b))
>> > > 1: brown(logflatnodes, archotreeouch)
>> > >
>> > > [[alternative HTML version deleted]]
>> > >
>> > > _______________________________________________
>> > > R-sig-phylo mailing list
>> > > R-sig-phylo@r-project.org
>> > > https://stat.ethz.ch/mailman/listinfo/r-sig-phylo
>
> --
> Aaron A. King, Ph.D.
> Ecology & Evolutionary Biology
> Mathematics
> Center for the Study of Complex Systems
> University of Michigan
> GPG Public Key: 0x15780975
>
_______________________________________________
R-sig-phylo mailing list
R-sig-phylo@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-phylo
