Benchmarks
Last updated: 2022-06-01
Checks: 6 1
Knit directory:
ampel-leipzig-meld/
This reproducible R Markdown analysis was created with workflowr (version 1.7.0). The Checks tab describes the reproducibility checks that were applied when the results were created. The Past versions tab lists the development history.
The R Markdown file has unstaged changes.
To know which version of the R Markdown file created these
results, you’ll want to first commit it to the Git repo. If
you’re still working on the analysis, you can ignore this
warning. When you’re finished, you can run
wflow_publish to commit the R Markdown file and
build the HTML.
Great job! The global environment was empty. Objects defined in the global environment can affect the analysis in your R Markdown file in unknown ways. For reproduciblity it’s best to always run the code in an empty environment.
The command set.seed(20210604) was run prior to running the code in the R Markdown file.
Setting a seed ensures that any results that rely on randomness, e.g.
subsampling or permutations, are reproducible.
Great job! Recording the operating system, R version, and package versions is critical for reproducibility.
Nice! There were no cached chunks for this analysis, so you can be confident that you successfully produced the results during this run.
Great job! Using relative paths to the files within your workflowr project makes it easier to run your code on other machines.
Great! You are using Git for version control. Tracking code development and connecting the code version to the results is critical for reproducibility.
The results in this page were generated with repository version 2f24e00. See the Past versions tab to see a history of the changes made to the R Markdown and HTML files.
Note that you need to be careful to ensure that all relevant files for the
analysis have been committed to Git prior to generating the results (you can
use wflow_publish or wflow_git_commit). workflowr only
checks the R Markdown file, but you know if there are other scripts or data
files that it depends on. Below is the status of the Git repository when the
results were generated:
Ignored files:
Ignored: _targets/
Ignored: container/
Ignored: logs/
Untracked files:
Untracked: 1907.00825.pdf
Untracked: METHODS
Untracked: TODO
Untracked: article.md
Untracked: s12874-018-0482-1.pdf
Untracked: scripts/R.sh
Untracked: scripts/env.sh
Unstaged changes:
Modified: _targets.R
Modified: analysis/article.Rmd
Modified: analysis/benchmarks.Rmd
Modified: analysis/bibliography/rpackages/article.bib
Modified: analysis/bibliography/rpackages/home.bib
Modified: analysis/elasticnet.Rmd
Modified: analysis/supplement.Rmd
Modified: code/01-data.R
Modified: code/05-ameld-elasticnet.R
Note that any generated files, e.g. HTML, png, CSS, etc., are not included in this status report because it is ok for generated content to have uncommitted changes.
These are the previous versions of the repository in which changes were made
to the R Markdown (analysis/benchmarks.Rmd) and HTML (docs/benchmarks.html)
files. If you’ve configured a remote Git repository (see
?wflow_git_remote), click on the hyperlinks in the table below to
view the files as they were in that past version.
| File | Version | Author | Date | Message |
|---|---|---|---|---|
| html | 983ec69 | Sebastian Gibb | 2022-03-17 | chore: rebuild site |
| Rmd | 057f935 | Sebastian Gibb | 2022-03-17 | feat: add elastic net bootstrap and timeROC evaluations |
| Rmd | d586eff | Sebastian Gibb | 2022-02-10 | feat: add benchmark table and plots |
| html | 373e7d8 | Sebastian Gibb | 2021-10-20 | chore: rebuild site |
| html | df8964f | Sebastian Gibb | 2021-10-15 | chore: rebuild site |
| Rmd | 0ac3045 | Sebastian Gibb | 2021-10-15 | refactor: reset figure width |
| Rmd | 173a1ca | Sebastian Gibb | 2021-10-14 | refactor: increase figure width |
| Rmd | 1e08cb2 | Sebastian Gibb | 2021-09-21 | fix: load bmrk_aggr and set p.value |
| Rmd | 939fcae | Sebastian Gibb | 2021-09-20 | feat: add benchmark comparison across datasets |
| Rmd | c66640c | Sebastian Gibb | 2021-09-14 | feat: first nnet tests |
| html | c66640c | Sebastian Gibb | 2021-09-14 | feat: first nnet tests |
| html | afa48d9 | Sebastian Gibb | 2021-08-07 | chore: rebuild site |
| Rmd | 4478f6a | Sebastian Gibb | 2021-08-02 | Revert "fix: working directory for targets" |
| Rmd | cac14f6 | Sebastian Gibb | 2021-08-02 | fix: working directory for targets |
| Rmd | 3957af7 | Sebastian Gibb | 2021-08-02 | refactor: move common yaml headers into _site.yml |
| html | 3aab3e1 | Sebastian Gibb | 2021-08-01 | chore: rebuild site |
| html | 3810a79 | Sebastian Gibb | 2021-07-15 | chore: rebuild site |
| Rmd | 1dc8b12 | Sebastian Gibb | 2021-07-14 | feat: add mlr benchmarks |
library("targets")
library("data.table")
library("mlr3")
library("mlr3misc")
library("mlr3proba")
library("mlr3viz")
library("viridis")
library("ggplot2")
tar_load(bmrk_results)
tar_load(bmrk_aggr)
agg <- bmrk_results$aggregate(msr("surv.cindex", id = "harrell"))
agg[, `:=`
(nr = NULL, resample_result = NULL, resampling_id = NULL, iters = NULL)
]
setorder(agg, task_id, -harrell)
p.value <- 0.051 Overview
1.1 Table
lapply(split(agg, agg$task_id), knitr::kable, digits = 4)$ln_eldd
| task_id | learner_id | harrell |
|---|---|---|
| ln_eldd | scale.ridge.tuned | 0.9212 |
| ln_eldd | scale.penridge.tuned | 0.9211 |
| ln_eldd | scale.svmregression.tuned | 0.9197 |
| ln_eldd | scale.elasticnet.tuned | 0.9195 |
| ln_eldd | scale.penlasso.tuned | 0.9135 |
| ln_eldd | ridge.tuned | 0.9132 |
| ln_eldd | penridge.tuned | 0.9129 |
| ln_eldd | ranger.tuned | 0.9119 |
| ln_eldd | scale.ranger.tuned | 0.9105 |
| ln_eldd | scale.surv.rfsrc.tuned | 0.9081 |
| ln_eldd | elasticnet.tuned | 0.9060 |
| ln_eldd | surv.rfsrc.tuned | 0.9054 |
| ln_eldd | penlasso.tuned | 0.9044 |
| ln_eldd | cox | 0.8632 |
| ln_eldd | scale.cox | 0.8552 |
| ln_eldd | scale.coxtime.tuned | 0.8472 |
| ln_eldd | scale.deepsurv.tuned | 0.8438 |
| ln_eldd | scale.surv.xgboost.tuned | 0.8368 |
| ln_eldd | surv.xgboost.tuned | 0.8365 |
| ln_eldd | lasso.tuned | 0.8335 |
| ln_eldd | scale.svmvanbelle1.tuned | 0.8331 |
| ln_eldd | scale.lasso.tuned | 0.8275 |
| ln_eldd | svmvanbelle1.tuned | 0.7794 |
| ln_eldd | deepsurv.tuned | 0.7494 |
| ln_eldd | coxtime.tuned | 0.7336 |
| ln_eldd | svmregression.tuned | 0.5711 |
$zlog_eldd
| task_id | learner_id | harrell |
|---|---|---|
| zlog_eldd | scale.ridge.tuned | 0.9211 |
| zlog_eldd | scale.svmregression.tuned | 0.9197 |
| zlog_eldd | scale.elasticnet.tuned | 0.9178 |
| zlog_eldd | scale.penridge.tuned | 0.9173 |
| zlog_eldd | ridge.tuned | 0.9154 |
| zlog_eldd | scale.penlasso.tuned | 0.9136 |
| zlog_eldd | elasticnet.tuned | 0.9133 |
| zlog_eldd | penridge.tuned | 0.9129 |
| zlog_eldd | penlasso.tuned | 0.9129 |
| zlog_eldd | ranger.tuned | 0.9120 |
| zlog_eldd | scale.ranger.tuned | 0.9113 |
| zlog_eldd | surv.rfsrc.tuned | 0.9074 |
| zlog_eldd | lasso.tuned | 0.9054 |
| zlog_eldd | scale.surv.rfsrc.tuned | 0.9042 |
| zlog_eldd | cox | 0.8619 |
| zlog_eldd | scale.cox | 0.8551 |
| zlog_eldd | scale.deepsurv.tuned | 0.8475 |
| zlog_eldd | scale.coxtime.tuned | 0.8462 |
| zlog_eldd | surv.xgboost.tuned | 0.8365 |
| zlog_eldd | scale.surv.xgboost.tuned | 0.8299 |
| zlog_eldd | scale.svmvanbelle1.tuned | 0.8267 |
| zlog_eldd | svmvanbelle1.tuned | 0.8190 |
| zlog_eldd | svmregression.tuned | 0.8093 |
| zlog_eldd | scale.lasso.tuned | 0.8007 |
| zlog_eldd | coxtime.tuned | 0.7994 |
| zlog_eldd | deepsurv.tuned | 0.7901 |
1.2 Boxplots
m <- agg[, max(harrell), by = task_id]
autoplot(bmrk_results) +
geom_boxplot(aes(fill = learner_id)) +
geom_jitter(position = position_jitter(0.2)) +
scale_fill_viridis(discrete = TRUE) +
geom_hline(
aes(yintercept = V1), linetype = "dashed", color = "red", data = m
)
autoplot(
bmrk_results,
measure = msr("surv.cindex", id = "uno", weight_meth = "G2")) +
geom_boxplot(aes(fill = learner_id)) +
geom_jitter(position = position_jitter(0.2)) +
scale_fill_viridis(discrete = TRUE)Warning: Removed 156 rows containing non-finite values (stat_boxplot).
Removed 156 rows containing non-finite values (stat_boxplot).Warning: Removed 156 rows containing missing values (geom_point).
| Version | Author | Date |
|---|---|---|
| 983ec69 | Sebastian Gibb | 2022-03-17 |
2 Aggregated performance
Aggregated performance across all 2 datasets.
autoplot(bmrk_aggr, type = "box", meas = "harrell") +
geom_boxplot(aes(fill = learner_id)) +
geom_jitter(position = position_jitter(0.2)) +
scale_fill_viridis(discrete = TRUE)
autoplot(bmrk_aggr, type = "box", meas = "uno") +
geom_boxplot(aes(fill = learner_id)) +
geom_jitter(position = position_jitter(0.2)) +
scale_fill_viridis(discrete = TRUE)Warning: Removed 50 rows containing non-finite values (stat_boxplot).
Removed 50 rows containing non-finite values (stat_boxplot).Warning: Removed 50 rows containing missing values (geom_point).
| Version | Author | Date |
|---|---|---|
| 983ec69 | Sebastian Gibb | 2022-03-17 |
3 Pairwise comparison
p-value is set to 0.05 (otherwise an error is thrown because of insignificant results).
autoplot(
bmrk_aggr, type = "fn", meas = "harrell", p.value = p.value,
col = palette.colors(2L)[2L]
)
4 Critical difference
p-value is set to 0.05 (otherwise an error is thrown because of insignificant results).
autoplot(
bmrk_aggr, type = "cd", meas = "harrell", style = 2, p.value = p.value,
minimize = FALSE
) + scale_color_viridis(discrete = TRUE)
sessionInfo()R version 4.1.3 (2022-03-10)
Platform: x86_64-unknown-linux-gnu (64-bit)
Matrix products: default
BLAS/LAPACK: /gnu/store/ras6dprsw3wm3swk23jjp8ww5dwxj333-openblas-0.3.18/lib/libopenblasp-r0.3.18.so
locale:
[1] C
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] ggplot2_3.3.5 viridis_0.6.2 viridisLite_0.4.0 mlr3viz_0.5.9
[5] mlr3proba_0.4.11 mlr3misc_0.10.0 mlr3_0.13.3 data.table_1.14.2
[9] targets_0.12.0
loaded via a namespace (and not attached):
[1] fs_1.5.2 bbotk_0.5.2 rprojroot_2.0.3
[4] mlr3pipelines_0.4.0 mlr3learners_0.5.2 tools_4.1.3
[7] backports_1.4.1 bslib_0.3.1 utf8_1.2.2
[10] R6_2.5.1 colorspace_2.0-3 withr_2.5.0
[13] tidyselect_1.1.2 gridExtra_2.3 processx_3.5.3
[16] compiler_4.1.3 git2r_0.30.1 cli_3.2.0
[19] ooplah_0.2.0 lgr_0.4.3 labeling_0.4.2
[22] bookdown_0.25 sass_0.4.1 BWStest_0.2.2
[25] scales_1.1.1 checkmate_2.0.0 mvtnorm_1.1-3
[28] callr_3.7.0 multcompView_0.1-8 palmerpenguins_0.1.0
[31] mlr3tuning_0.13.0 stringr_1.4.0 digest_0.6.29
[34] mlr3extralearners_0.5.9 rmarkdown_2.13 param6_0.2.4
[37] paradox_0.8.0 set6_0.2.4 pkgconfig_2.0.3
[40] htmltools_0.5.2 parallelly_1.31.0 mlr3benchmark_0.1.3
[43] fastmap_1.1.0 highr_0.9 rlang_1.0.2
[46] SuppDists_1.1-9.7 jquerylib_0.1.4 generics_0.1.2
[49] farver_2.1.0 jsonlite_1.8.0 dplyr_1.0.8
[52] magrittr_2.0.3 Matrix_1.4-1 Rcpp_1.0.8.3
[55] munsell_0.5.0 fansi_1.0.3 lifecycle_1.0.1
[58] stringi_1.7.6 whisker_0.4 yaml_2.3.5
[61] MASS_7.3-56 grid_4.1.3 parallel_4.1.3
[64] dictionar6_0.1.3 listenv_0.8.0 promises_1.2.0.1
[67] crayon_1.5.1 lattice_0.20-45 PMCMRplus_1.9.4
[70] splines_4.1.3 knitr_1.38 ps_1.6.0
[73] pillar_1.7.0 igraph_1.3.0 uuid_1.0-4
[76] base64url_1.4 kSamples_1.2-9 codetools_0.2-18
[79] glue_1.6.2 evaluate_0.15 vctrs_0.4.0
[82] httpuv_1.6.5 distr6_1.6.9 gtable_0.3.0
[85] purrr_0.3.4 future_1.24.0 cachem_1.0.6
[88] xfun_0.30 Rmpfr_0.8-7 later_1.3.0
[91] survival_3.3-1 tibble_3.1.6 memoise_2.0.1
[94] workflowr_1.7.0 gmp_0.6-5 globals_0.14.0
[97] ellipsis_0.3.2