search

pbreheny / biglasso

biglasso: Extending Lasso Model Fitting to Big Data in R
http://pbreheny.github.io/biglasso/
113 stars 29 forks source link

predict.cv.biglasso - address 0x2b2da5f69ae0, cause 'memory not mapped' #30

Closed prateeksasan1 closed 3 years ago

prateeksasan1 commented 4 years ago

I was trying to predict.cv.biglasso and got the following error :

I havent been able to figure why this error occurs. Any ideas?

caught segfault address 0x2b2da5f69ae0, cause 'memory not mapped'

Traceback: 1: get_eta(X@address, as.integer(row.idx - 1), beta, beta.T@i, beta.T@j) 2: predict.biglasso(object$fit, X = X, row.idx = row.idx, type = type, lambda = lambda, which = which, ...) 3: predict.cv.biglasso(fit1, X, type = "response", lambda = fit1$lambda[fit1$min]) 4: predict(fit1, X, type = "response", lambda = fit1$lambda[fit1$min]) 5: as.vector(predict(fit1, X, type = "response", lambda = fit1$lambda[fit1$min])) 6: lasso_reg(test_sub, "101_anat_fmri.nii.gz", "afni_biglasso_selection_cv_fs+orig.BRIK.gz", "penalty_bglasso_obj_fs.RData", "fs") An irrecoverable exception occurred. R is aborting now ... /var/spool/torque/mom_priv/jobs/11529146.owens-batch.ten.osc.edu.SC: line 43: 7338 Segmentation fault Rscript /users/PAS1316/prateeksasan/hcp/wexner/controls_lasso_reg_oos.R

pbreheny commented 4 years ago

Can you provide a reproducible example? This is impossible to debug if we can't see what's causing the error.

privefl commented 4 years ago

I think it would be useful to know e.g. which type of big.matrix object you are using and if e.g. you store it in some temp directory. I remember having a similar issue when that was the case.