Error in check.data.partitions.covmats(args)

Question

Error in check.data.partitions.covmats(args)

Closed this issue 4 years ago · 8 comments

Hi,

I try to run the cross validation analysis the following error pop up:
Error in check.data.partitions.covmats(args) :
you have specified an invalid data partition "data" element is not positive definite.

these the code:
my.xvals <- x.validation(train.prop = 0.9, n.reps = 10, K = 1:5,
freqs = freqs,
data.partitions = NULL,
geoDist = geoDist, coords = coords,
prefix = "Crau_K1_5", n.iter = 1000,
make.figs = FALSE, save.files = TRUE,
parallel = FALSE, n.nodes = 1)

Please, can you help what is going on error? thank you in advance,

r

Answer 1 · 2019-08-02T19:00:19.000Z

Hi Ruth,

Please see #20 (comment) or #21 (comment) about dealing with nonpositive-definite issues. If that doesn't resolve it, please attach or email your allele frequency data as a .txt data table and I'll take a look.

-Gideon

Answer 2 · 2019-08-02T20:00:27.000Z

Hi Gideon, Here is my freqs file. I think I have a lot missing data and maybe it can be the reason of the error. Here is my file, Thank you in advance, r El vie., 2 ago. 2019 a las 15:00, gbradburd (<notifications@github.com>) escribió:

…

Hi Ruth, Please see #20 (comment) <#20 (comment)> or #21 (comment) <#21 (comment)> about dealing with nonpositive-definite issues. If that doesn't resolve it, please attach or email your allele frequency data as a .txt data table and I'll take a look. -Gideon — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#23?email_source=notifications&email_token=AEL3NLLTYGBJF73T7FTGYLLQCR74JA5CNFSM4II7WDK2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD3OSSAA#issuecomment-517810432>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AEL3NLI6ELOEYKPRTC34ROTQCR74JANCNFSM4II7WDKQ> .

--

Answer 3 · 2019-08-02T20:09:01.000Z

Looks like the file failed to attach.

Answer 4 · 2019-08-02T20:11:01.000Z

here is the file
freqs.txt

Answer 5 · 2019-08-02T20:41:13.000Z

Hi Ruth,

Yes, it looks like the problem is lots of missing data. Your dataset appears to have 49% missing data total, and, of your 64 samples, 28 have greater than 50% missing data. I would recommend one of the following approaches:

try dropping loci and/or samples with large amounts of missing data until conStruct runs for you.
it looks like your samples are in different populations (e.g., "ArrNeg1"..."ArrNeg7", "CINvLi1"..."CINvLi8"), so, depending on the spatial scale of the sampling, you could try collapsing some of the individuals from the same population into a single, multi-individual sample in the hope that it would ameliorate some of your missing data problems. I describe that procedure in #21 (comment).

Hope that helps,
-Gideon

Answer 6 · 2019-08-02T20:46:30.000Z

Hi Gideon,
Thanks, I gonna try if and I'll see if that works! Thanks!,
r

Answer 7 · 2020-08-27T12:52:01.000Z

Hi Ruth - is this issue still unresolved for you? If not, I'll go ahead and close it out.

Answer 8 · 2020-08-27T19:13:21.000Z

I did it, Thank you! r El jue., 27 ago. 2020 a las 7:52, Gideon (<notifications@github.com>) escribió:

…

Hi Ruth - is this issue still unresolved for you? If not, I'll go ahead and close it out. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#23 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AEL3NLNOKXF5DUA6KS42YPLSCZJIBANCNFSM4II7WDKQ> .