Проблема
Я использую ggwithinplot в ggstatsplot для построения 1755 точек данных, и она оказывается очень медленной.
Я подтвердил эту проблему, протестировав затраченное время на построение разного количества точек данных:
1.N = 120, 5,576425 с.
2.N = 360, 11,90681 с.
3.N = 1755, 6,699168 минут , но не секунд.
Что я пробовал
- Я преобразовал data.frame в data.table. Это помогло.
- Я установил point.path = FALSE в ggwithinplot. Это помогло.
- Я установил nboot = 1 в ggwithinplot, чтобы уменьшить время для расчета CI величины эффекта. Это помогло.
Код
ggwithinstats(long_general_disBU_dt, x = condition, y = disBU,
point.path = FALSE,
mean.ci = TRUE,
pairwise.comparisons = TRUE,
pairwise.annotation = "asterisk",
pairwise.display = "s",
effsize.type = 'partial_eta',
p.adjust.method = "fdr",
ggtheme = theme_classic(),
palette = "Darjeeling2",
package = "wesanderson",
ggstatsplot.layer = FALSE,
bf.message= FALSE,
nboot=1,
xlab = "targets",
ylab = "general_disBU_dt",
title = "general disBU across 20 news")
информация о сеансе:
R version 3.6.3 (2020-02-29)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 15063)
Matrix products: default
locale:
[1] LC_COLLATE=Chinese (Simplified)_China.936 LC_CTYPE=Chinese (Simplified)_China.936
[3] LC_MONETARY=Chinese (Simplified)_China.936 LC_NUMERIC=C
[5] LC_TIME=Chinese (Simplified)_China.936
attached base packages:
[1] splines stats4 stats graphics grDevices utils datasets methods base
other attached packages:
[1] reshape_0.8.8 gginnards_0.0.3 VGAM_1.1-2 parameters_0.6.0 nnet_7.3-13 openxlsx_4.1.4
[7] summarytools_0.9.6 ggcorrplot_0.1.3 bruceR_0.4.0 performance_0.4.4 lubridate_1.7.4 psych_1.9.12.31
[13] data.table_1.12.8 rio_0.5.16 ggstatsplot_0.3.1 forcats_0.5.0 stringr_1.4.0 dplyr_0.8.5
[19] purrr_0.3.3 readr_1.3.1 tidyr_1.0.2 tibble_2.1.3 ggplot2_3.3.0 tidyverse_1.3.0
[25] drawMap_0.1.0
loaded via a namespace (and not attached):
[1] estimability_1.3 GGally_1.4.0 lavaan_0.6-5 coda_0.19-3
[5] acepack_1.4.1 knitr_1.28 multcomp_1.4-12 rpart_4.1-15
[9] inline_0.3.15 generics_0.0.2 callr_3.4.2 cowplot_1.0.0
[13] TH.data_1.0-10 xml2_1.2.5 httpuv_1.5.2 StanHeaders_2.21.0-1
[17] assertthat_0.2.1 d3Network_0.5.2.1 WRS2_1.0-0 xfun_0.12
[21] hms_0.5.3 promises_1.1.0 fansi_0.4.1 dbplyr_1.4.2
[25] readxl_1.3.1 igraph_1.2.4.2 htmlwidgets_1.5.1 DBI_1.1.0
[29] Rsolnp_1.16 paletteer_1.1.0 rcompanion_2.3.25 backports_1.1.5
[33] pbivnorm_0.6.0 insight_0.8.2 rapportools_1.0 libcoin_1.0-5
[37] jmvcore_1.2.5 vctrs_0.2.4 sjlabelled_1.1.3 abind_1.4-5
[41] withr_2.1.2 pryr_0.1.4 metaBMA_0.6.2 checkmate_2.0.0
[45] bdsmatrix_1.3-4 emmeans_1.4.5 fdrtool_1.2.15 prettyunits_1.1.1
[49] fastGHQuad_1.0 mnormt_1.5-6 cluster_2.1.0 mi_1.0
[53] crayon_1.3.4 pkgconfig_2.0.3 nlme_3.1-145 statsExpressions_0.3.1
[57] palr_0.2.0 pals_1.6 rlang_0.4.5 lifecycle_0.2.0
[61] miniUI_0.1.1.1 groupedstats_0.2.0 skimr_2.1 LaplacesDemon_16.1.4
[65] MatrixModels_0.4-1 sandwich_2.5-1 kutils_1.69 EMT_1.1
[69] modelr_0.1.6 dichromat_2.0-0 tcltk_3.6.3 cellranger_1.1.0
[73] matrixStats_0.56.0 broomExtra_2.5.0 lmtest_0.9-37 Matrix_1.2-18
[77] regsem_1.5.2 loo_2.2.0 mc2d_0.1-18 carData_3.0-3
[81] boot_1.3-24 zoo_1.8-7 reprex_0.3.0 base64enc_0.1-3
[85] whisker_0.4 processx_3.4.2 png_0.1-7 viridisLite_0.3.0
[89] rjson_0.2.20 oompaBase_3.2.9 pander_0.6.3 ggExtra_0.9
[93] afex_0.26-0 multcompView_0.1-8 coin_1.3-1 arm_1.10-1
[97] jpeg_0.1-8.1 rockchalk_1.8.144 ggsignif_0.6.0 scales_1.1.0
[101] magrittr_1.5 plyr_1.8.6 compiler_3.6.3 rstantools_2.0.0
[105] bbmle_1.0.23.1 RColorBrewer_1.1-2 lme4_1.1-21 cli_2.0.2
[109] lmerTest_3.1-1 pbapply_1.4-2 ps_1.3.2 TMB_1.7.16
[113] Brobdingnag_1.2-6 htmlTable_1.13.3 Formula_1.2-3 MASS_7.3-51.5
[117] mgcv_1.8-31 tidyselect_1.0.0 stringi_1.4.6 lisrelToR_0.1.4
[121] sem_3.1-9 jtools_2.0.2 OpenMx_2.17.3 latticeExtra_0.6-29
[125] ggrepel_0.8.2 bridgesampling_1.0-0 grid_3.6.3 tools_3.6.3
[129] parallel_3.6.3 matrixcalc_1.0-3 rstudioapi_0.11 foreign_0.8-76
[133] gridExtra_2.3 ipmisc_1.2.0 pairwiseComparisons_0.2.5 BDgraph_2.62
[137] digest_0.6.25 shiny_1.4.0.2 nortest_1.0-4 jmv_1.2.5
[141] Rcpp_1.0.3 car_3.0-7 broom_0.5.5 metafor_2.1-0
[145] ez_4.4-0 BayesFactor_0.9.12-4.2 metaplus_0.7-11 later_1.0.0
[149] httr_1.4.1 effectsize_0.2.0 sjstats_0.17.9 colorspace_1.4-1
[153] rvest_0.3.5 XML_3.99-0.3 fs_1.3.2 truncnorm_1.0-8
[157] rematch2_2.1.0 expm_0.999-4 mapproj_1.2.7 jcolors_0.0.4
[161] MuMIn_1.43.15 xtable_1.8-4 jsonlite_1.6.1 nloptr_1.2.2.1
[165] corpcor_1.6.9 rstan_2.19.3 glasso_1.11 zeallot_0.1.0
[169] modeltools_0.2-23 scico_1.1.0 R6_2.4.1 Hmisc_4.3-1
[173] broom.mixed_0.2.4 pillar_1.4.3 htmltools_0.4.0 mime_0.9
[177] glue_1.3.2 fastmap_1.0.1 minqa_1.2.4 codetools_0.2-16
[181] maps_3.3.0 pkgbuild_1.0.6 mvtnorm_1.1-0 lattice_0.20-40
[185] numDeriv_2016.8-1.1 huge_1.3.4 curl_4.3 DescTools_0.99.34
[189] gtools_3.8.1 magick_2.3 logspline_2.1.15 zip_2.0.4
[193] survival_3.1-11 qgraph_1.6.5 repr_1.1.0 munsell_0.5.0
[197] semPlot_1.1.2 sjmisc_2.8.3 haven_2.2.0 reshape2_1.4.3
[201] gtable_0.3.0 bayestestR_0.5.2
Могу ли я что-нибудь сделать для ускорить функцию ggwithinplot?