-
Notifications
You must be signed in to change notification settings - Fork 19
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Progress Bar for GLMM #331
Comments
Resource use is controlled by a combination of OpenMP and BiocParallel - the latter for parallelising over the nhoods, and OpenMP for multi-threading in the Rcpp code. A default for GLMMs tend to scale poorly - while I have implemented a range of measures to cut this down, i.e. parallelisation and multi-threading, however, the bottlenecks are several dense matrix multiplications which are ~ I'm not sure why your Rstudio is ignoring environmental variables, as this is the recommendation: https://cran.r-project.org/doc/manuals/r-release/R-exts.html#OpenMP-support - you may need to seek Rstudio support to resolve the issue. |
Specifying sample ID as a fixed effect isn't a quick workaround to avoid enriched neighbourhoods comprised of one sample. design <- data.frame(colData(MILOdata))[, c("Sample", "Smoker")]
design <- distinct(design)
rownames(design) <- design$Sample
testNhoods(MILOdata, design = ~ Smoker + Sample, design.df = design, fdr.weighting = "graph-overlap")
Using TMM normalisation
Error in glmFit.default(y, design = design, dispersion = dispersion, offset = offset, :
Design matrix not of full rank. The following coefficients not estimable:
SampleOSCC_26-P |
Hi @MikeDMorgan, I performed a GLMM analysis. However, it threw an error that I couldn't identify. Is there any way to check where the error is coming from in this situation? |
Parameter |
Here is another problem: after performing the GLMM, my DA results include NA values. However,
|
@yulijia You can use On a separate note - your issue is unrelated to the OP, so you should open a new issue not append yours to the end of this one. |
Is it feasible to add a progress bar if a mixed model is used? It has been running for the past five hours without a progress message.
It also appears to consume all CPUs on a server, causing CPU starvation due to RStudio Server ignoring a user's
~/.bashrc
file with:I wonder how to choose between different options for
glmm.solver
in case one of them is less resource-intensive. I tried"Fisher"
.The text was updated successfully, but these errors were encountered: