Possible incompatibility between pec and redomForestSRC:::rfs during package upgrading

I am trying to use the package: https://cran.r-project.org/web/packages/randomForestSRC/randomForestSRC.pdf to do error estimate my random survival forest models.

If I run rfsrc on my data, I get correct result. If I run pec for my cox model, I also get correct result. But if I run pec for rfsrc model, and if I set B=50, it stuck at 20 and do not progress. If Iset B=10 or 20, it goes through full 10 or 20, but stops there and does not finish.

I also tried with the data in the reference, and same things happened. The packages are all installed very recently, in the past month or so.

sessionInfo() R version 4.1.3 (2022-03-10) Platform: x86_64-pc-linux-gnu (64-bit) Running under: CentOS Linux 7 (Core)

Matrix products: default BLAS: /home/projects/cu_00010/people/haikon/Software/R-4.1.3/lib/libRblas.so LAPACK: /home/projects/cu_00010/people/haikon/Software/R-4.1.3/lib/libRlapack.so

locale: [1] LC_CTYPE=en_US.utf-8 LC_NUMERIC=C [3] LC_TIME=en_US.utf-8 LC_COLLATE=en_US.utf-8 [5] LC_MONETARY=en_US.utf-8 LC_MESSAGES=en_US.utf-8 [7] LC_PAPER=en_US.utf-8 LC_NAME=C [9] LC_ADDRESS=C LC_TELEPHONE=C [11] LC_MEASUREMENT=en_US.utf-8 LC_IDENTIFICATION=C

attached base packages: [1] parallel stats graphics grDevices utils datasets methods [8] base

other attached packages: [1] doMC_1.3.8 iterators_1.0.14 [3] foreach_1.5.2 riskRegression_2022.03.22 [5] pec_2022.05.04 randomForestSRC_3.1.1 [7] prodlim_2019.11.13 survival_3.4-0 [9] data.table_1.14.2

loaded via a namespace (and not attached): [1] jsonlite_1.8.0 splines_4.1.3 Formula_1.2-4 [4] assertthat_0.2.1 latticeExtra_0.6-30 globals_0.16.1 [7] timereg_2.0.2 numDeriv_2016.8-1.1 pillar_1.8.1 [10] backports_1.4.1 lattice_0.20-45 quantreg_5.94 [13] glue_1.6.2 digest_0.6.29 RColorBrewer_1.1-3 [16] checkmate_2.1.0 sandwich_3.0-2 colorspace_2.0-3 [19] cmprsk_2.2-11 rms_6.3-0 htmltools_0.5.3 [22] Matrix_1.4-1 pkgconfig_2.0.3 SparseM_1.81 [25] listenv_0.8.0 DiagrammeR_1.0.9 purrr_0.3.4 [28] mvtnorm_1.1-3 scales_1.2.1 jpeg_0.1-9 [31] lava_1.6.10 MatrixModels_0.5-0 tibble_3.1.7 [34] htmlTable_2.4.1 mets_1.3.0 generics_0.1.3 [37] ggplot2_3.3.6 ellipsis_0.3.2 TH.data_1.1-1 [40] nnet_7.3-17 cli_3.3.0 magrittr_2.0.3 [43] deldir_1.0-6 polspline_1.1.20 data.tree_1.0.0 [46] future_1.28.0 fansi_1.0.3 parallelly_1.32.1 [49] nlme_3.1-159 MASS_7.3-58.1 foreign_0.8-82 [52] tools_4.1.3 multcomp_1.4-20 lifecycle_1.0.1 [55] stringr_1.4.1 interp_1.1-3 munsell_0.5.0 [58] cluster_2.1.4 compiler_4.1.3 rlang_1.0.3 [61] grid_4.1.3 rstudioapi_0.14 htmlwidgets_1.5.4 [64] visNetwork_2.1.0 base64enc_0.1-3 gtable_0.3.1 [67] codetools_0.2-18 DBI_1.1.3 R6_2.5.1 [70] zoo_1.8-10 gridExtra_2.3 knitr_1.40 [73] dplyr_1.0.9 fastmap_1.1.0 future.apply_1.9.1 [76] utf8_1.2.2 Hmisc_4.7-1 stringi_1.7.8 [79] Rcpp_1.0.9 vctrs_0.4.1 rpart_4.1.16 [82] png_0.1-7 tidyselect_1.1.2 xfun_0.32

scores = Score(object=mod, formula=Surv(Time,Status)~1, data=data, B=10, M=600,

summary=c('risks','riskQuantile','ibs'), plots='Calibration',

times=c(2,5,8,10)*365.25, null.model=FALSE, conf.int=TRUE, conrast=TRUE,

split.method='bootcv', parallel='as.registered', ncpus=20, fill=TRUE) Running crossvalidation algorithm Fitting the models in 10 learning datasets, then predicting the risks in validation datasets |====================| 100%

scores = Score(object=mod, formula=Surv(Time,Status)~1, data=data, B=200, M=n.subsample,

summary=c('risks','riskQuantile','ibs'), plots='Calibration',

times=c(2,5,8,10)*365.25, null.model=FALSE, conf.int=TRUE, conrast=TRUE,

split.method='bootcv', parallel='as.registered', ncpus=n.cpu, fill=TRUE) Running crossvalidation algorithm Fitting the models in 200 learning datasets, then predicting the risks in validation datasets |== | 10%^C

Please help me so that i could finish my project.

kogalur / randomForestSRC

Possible incompatibility between pec and redomForestSRC:::rfs during package upgrading #330