layer_morphospace_LDA issue

[lrocax]

Hi,

What I do wrong with the layer_morphospace_LDA script as I get this error:
Error in x$mod.pred$x[, axes] : subscript out of bounds

library(Momocs)

lf <- list.files("C:/Data/outline/", pattern = ".txt$",full.names=TRUE)
lf1<-lf_structure(lf, split = "C:/ Data /outline/", trim.extension = FALSE)
lf2<-data.frame(substr(lf1$V2, 1, 5))
names(lf2)[1] <- "Type"

coordinates <- import_txt(lf)
do <- Out(coordinates, fac = lf2)

do.f <- efourier(do, nb.h=32, norm=TRUE)

do.p <- PCA(do.f)

do.l <- LDA(do.p, ~Type)
plot_LDA(do.l)
plot_LDA(do.l, 'Type') + layer_morphospace_LDA(do.l)

[vbonhomme]

I m on holidays and afk but i gonna have a look asap. Le sam. 20 févr. 2021 15:39, Lluis Roca <notifications@github.com> a écrit :

…

Hi, What I do wrong with the layer_morphospace_LDA script as I get this error: *Error in x$mod.pred$x[, axes] : subscript out of bounds* library(Momocs) lf <- list.files("C:/Data/outline/", pattern = ".txt$",full.names=TRUE) lf1<-lf_structure(lf, split = "C:/ Data /outline/", trim.extension = FALSE) lf2<-data.frame(substr(lf1$V2, 1, 5)) names(lf2)[1] <- "Type" coordinates <- import_txt(lf) do <- Out(coordinates, fac = lf2) do.f <- efourier(do, nb.h=32, norm=TRUE) do.p <- PCA(do.f) do.l <- LDA(do.p, ~Type) plot_LDA(do.l) plot_LDA(do.l, 'Type') + layer_morphospace_LDA(do.l) — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#218>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABECQFSAD4VK2ZK24C2ZBBLS77CRVANCNFSM4X553KPQ> .

[vbonhomme]

Does type only have two levels ? If so this cannot work :( Le sam. 20 févr. 2021 15:39, Lluis Roca <notifications@github.com> a écrit :

…

Hi, What I do wrong with the layer_morphospace_LDA script as I get this error: *Error in x$mod.pred$x[, axes] : subscript out of bounds* library(Momocs) lf <- list.files("C:/Data/outline/", pattern = ".txt$",full.names=TRUE) lf1<-lf_structure(lf, split = "C:/ Data /outline/", trim.extension = FALSE) lf2<-data.frame(substr(lf1$V2, 1, 5)) names(lf2)[1] <- "Type" coordinates <- import_txt(lf) do <- Out(coordinates, fac = lf2) do.f <- efourier(do, nb.h=32, norm=TRUE) do.p <- PCA(do.f) do.l <- LDA(do.p, ~Type) plot_LDA(do.l) plot_LDA(do.l, 'Type') + layer_morphospace_LDA(do.l) — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#218>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABECQFSAD4VK2ZK24C2ZBBLS77CRVANCNFSM4X553KPQ> .

[lrocax]

I have in the dataset three groups (If you mean two levels as two groups than)

[vbonhomme]

Ok. Gonna have a look next week Le dim. 21 févr. 2021 20:57, Lluis Roca <notifications@github.com> a écrit :

…

I have in the dataset three groups (If you mean two levels as two groups than) — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#218 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABECQFS26FVRXSSWRUYIHMTTAFQSRANCNFSM4X553KPQ> .

[vbonhomme]

Have you tried : plot_LDA(do.l, morphospace=T) Or plot_LDA(do.l) + layer_morphospace_LDA() ? Wih plot_LDA you dont have to mention the fac, as its already defined in the LDA call. Le lun. 22 févr. 2021 11:52, Vincent Bonhomme <bonhomme.vincent@gmail.com> a écrit :

…

Ok. Gonna have a look next week Le dim. 21 févr. 2021 20:57, Lluis Roca ***@***.***> a écrit : > I have in the dataset three groups (If you mean two levels as two groups > than) > > — > You are receiving this because you commented. > Reply to this email directly, view it on GitHub > <#218 (comment)>, or > unsubscribe > <https://github.com/notifications/unsubscribe-auth/ABECQFS26FVRXSSWRUYIHMTTAFQSRANCNFSM4X553KPQ> > . >

[lrocax]

I tried plot_LDA(do.l) + layer_morphospace_LDA() and I got this error:

• layer_morphospace_LDA is back, but experimental
  Error in layer_morphospace_LDA() :
  argument "x" is missing, with no default

[vbonhomme]

What about plot_LDA(do.l, morphospace=T) ? Le ven. 5 mars 2021 22:05, Lluis Roca <notifications@github.com> a écrit :

…

I tried plot_LDA(do.l) + layer_morphospace_LDA() and I got this error: - layer_morphospace_LDA is back, but experimental Error in layer_morphospace_LDA() : argument "x" is missing, with no default — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#218 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABECQFQRY77G5PLEEX73Z3LTCFBRLANCNFSM4X553KPQ> .

[lrocax]

I also got an error:

• layer_morphospace_LDA is back, but experimental
  Error in apply(coo, 2, function(x) diff(range(x))) :
  dim(X) must have a positive length