transfer from old repo

2 Longterm/260315/_targets.R

@@ -0,0 +1,285 @@
+# Created by use_targets().
+# Follow the comments below to fill in this target script.
+# Then follow the manual to check and run the pipeline:
+#   https://books.ropensci.org/targets/walkthrough.html#inspect-the-pipeline
+
+# Load packages required to define the pipeline:
+library(targets)
+library(tarchetypes) # Load other packages as needed.
+
+# Set target options:
+tar_option_set(
+  seed = 142,
+  packages = c("tibble"), # packages that your targets need to run
+  format = "qs" # , # Optionally set the default storage format. qs is fast.
+  #
+  # For distributed computing in tar_make(), supply a {crew} controller
+  # as discussed at https://books.ropensci.org/targets/crew.html.
+  # Choose a controller that suits your needs. For example, the following
+  # sets a controller with 2 workers which will run as local R processes:
+  #
+  #   controller = crew::crew_controller_local(workers = 2)
+  #
+  # Alternatively, if you want workers to run on a high-performance computing
+  # cluster, select a controller from the {crew.cluster} package. The following
+  # example is a controller for Sun Grid Engine (SGE).
+  #
+  #   controller = crew.cluster::crew_controller_sge(
+  #     workers = 50,
+  #     # Many clusters install R as an environment module, and you can load it
+  #     # with the script_lines argument. To select a specific verison of R,
+  #     # you may need to include a version string, e.g. "module load R/4.3.0".
+  #     # Check with your system administrator if you are unsure.
+  #     script_lines = "module load R"
+  #   )
+  #
+  # Set other options as needed.
+)
+
+# tar_make_clustermq() is an older (pre-{crew}) way to do distributed computing
+# in {targets}, and its configuration for your machine is below.
+options(clustermq.scheduler = "multiprocess")
+
+# tar_make_future() is an older (pre-{crew}) way to do distributed computing
+# in {targets}, and its configuration for your machine is below.
+future::plan(future.callr::callr)
+
+# Run the R scripts in the R/ folder with your custom functions:
+tar_source()
+source(here::here("R/glmnet-reg.R")) # Source other scripts as needed.
+
+# Replace the target list below with your own:
+list(
+  tar_target(
+    name = pop_df,
+    command = get_clinical()
+  ),
+  tar_target(
+    name = reg_list,
+    command = get_reg_ls()
+  ),
+  tar_target(
+    name = df_deaths,
+    command = get_deaths(reg_list)
+  ),
+  tar_target(
+    name = df_events,
+    command = get_events(reg_list)
+  ),
+  tar_target(
+    name = df_events_deaths,
+    command = merge_events(list(events = df_events, deaths = df_deaths, clinical = pop_df))
+  ),
+  tar_target(
+    name = df_treated,
+    command = get_treated(reg_list)
+  ),
+  tar_target(
+    name = df_dst,
+    command = get_dst(reg_list, pop_df)
+  ),
+  tar_target(
+    name = list_filtered,
+    command = list(all_events = df_events_deaths, clinical = pop_df, dst = df_dst)
+  ),
+  tar_target(
+    name = df_all_data,
+    command = collectall(list_filtered)
+  ),
+  tar_target(
+    name = df_all_data_formatted,
+    command = data_formatting(df_all_data)
+  ),
+  tar_target(
+    name = ls_all_events,
+    command = all_events(list(events = df_events, deaths = df_deaths, clinical = pop_df))
+  ),
+  tar_target(
+    name = df_event_data,
+    command = events_ready(df_all_data_formatted)
+  ),
+  tar_target(
+    name = df_talos_data,
+    command = talos_ready(df_all_data_formatted)
+  ),
+  tar_target(
+    name = df_talos_data_imp,
+    command = talos_imp(df_talos_data)
+  ),
+  tar_target(
+    name = tbl_events_summary,
+    command = events_tblone(df_event_data)
+  ),
+  tar_target(
+    name = tbl_events_cox_regression,
+    command = show_table_regression(df_event_data, use.mice = FALSE)
+  ),
+  tar_target(
+    name = tbl_events_cox_regression_uv,
+    command = uv_cox_table(df_event_data)
+  ),
+  tar_target(
+    name = df_event_data_small,
+    command = events_ready_small(df_all_data_formatted)
+  ),
+  tar_target(
+    name = tbl_events_summary_small,
+    command = events_tblone(df_event_data_small)
+  ),
+  tar_target(
+    name = tbl_events_cox_regression_small,
+    command = show_table_regression(df_event_data_small, use.mice = FALSE)
+  ),
+  tar_target(
+    name = df_events_mids,
+    command = events_dataset(df_event_data, impute = TRUE)
+  ),
+  tar_target(
+    name = df_events_complete,
+    command = complete_preds_data(df_event_data)
+  ),
+  tar_target(
+    name = tbl_events_mids_cox_regression,
+    command = show_table_regression(df_event_data, use.mice = TRUE)
+  ),
+  tar_target(
+    name = plot_events_survival_smooth,
+    command = df_event_data |> events_dataset(impute = FALSE) |> cox_regression(include_formula=TRUE) |> plot_survival_smooth()
+  )#,
+  # tar_target(
+  #   name = plot_events_survival_smooth_mids,
+  #   command = df_events_mids |> cox_regression() |> plot_survival_smooth()
+  # ),
+  # tar_target(
+  #   name = df_pred_data,
+  #   command = prediction_ready(df_all_data_formatted)
+  # ),
+  # tar_target(
+  #   name = tbl_pred_summary,
+  #   command = preds_tblone(df_pred_data)
+  # ),
+  # tar_target(
+  #   name = tbl_pred_summary_true,
+  #   command = df_pred_data |>
+  #     dplyr::select(pase_0, pase_4, soc_status_nowork, fam_indk_hl, edu_level_hl) |>
+  #     true_pred_sum_plot()
+  # ),
+  # tar_target(
+  #   name = tbl_pred_summary_exp,
+  #   command = df_pred_data |>
+  #     dplyr::select(pase_0, pase_4, soc_status_nowork, fam_indk_hl, edu_level_hl) |>
+  #     preds_tblone()
+  # ),
+  # tar_target(
+  #   name = tbl_pred_summary_exp_sex,
+  #   command = df_pred_data |>
+  #     dplyr::select(reg_female, soc_status_nowork, fam_indk_hl, edu_level_hl) |>
+  #     summary_tblone()
+  # ),
+  # tar_target(
+  #   name = tbl_pred_summary_exp_pase,
+  #   command = df_pred_data |> sum_pase_tables()
+  # ),
+  # tar_target(
+  #   name = tbl_pred_summary_exp_missing_edu,
+  #   command = df_pred_data |>
+  #     dplyr::select(age,reg_female, reg_trombolyse, reg_trombektomi, reg_hyperten, reg_diabetes, nihss_0, pase_0, pase_4, soc_status_nowork, fam_indk_hl, edu_level_hl) |>
+  #     who_is_missing(var="edu_level_hl")
+  # ),
+  # tar_target(
+  #   name = tbl_pred_summary_exp_missing_bmi,
+  #   command = df_pred_data |>
+  #     dplyr::select(age,reg_female, reg_trombolyse, reg_trombektomi, reg_hyperten, reg_diabetes, nihss_0, pase_0, pase_4, reg_bmi, soc_status_nowork, fam_indk_hl, edu_level_hl) |>
+  #     who_is_missing(var="reg_bmi")
+  # ),
+  # tar_target(
+  #   name = ls_pred_models,
+  #   command = pred_models(df_pred_data,auto.l = TRUE,weighted = FALSE)
+  # ),
+  # tar_target(
+  #   name = ls_pred_models_anyupdown,
+  #   command = pred_models(df_pred_data,auto.l = TRUE,weighted = FALSE,split.type="anyupdown")
+  # ),
+  # tar_target(
+  #   name = ls_pred_models_relupdown_20,
+  #   command = pred_models(df_pred_data,auto.l = TRUE,weighted = FALSE,split.type="relupdown",rel.bin=20)
+  # ),
+  # tar_target(
+  #   name = ls_pred_models_relupdown_50,
+  #   command = pred_models(df_pred_data,auto.l = TRUE,weighted = FALSE,split.type="relupdown",rel.bin=50)
+  # ),
+  # tar_target(
+  #   name = df_pred_mids,
+  #   command = fun_impute(data = df_pred_data, outcome.vars = c("pase_0", "pase_4"))
+  # ),
+  # tar_target(
+  #   name = ls_pred_mids_reg,
+  #   command = mids_regularisation(df_pred_mids)
+  # ),
+  # tar_target(
+  #   name = ls_pred_mids_reg_anyupdown,
+  #   command = mids_regularisation(df_pred_mids,split.type="anyupdown")
+  # ),
+  # tar_target(
+  #   name = ls_pred_mids_reg_relupdown_20,
+  #   command = mids_regularisation(df_pred_mids,split.type="relupdown",rel.bin=20)
+  # ),
+  # tar_target(
+  #   name = ls_pred_mids_reg_relupdown_50,
+  #   command = mids_regularisation(df_pred_mids,split.type="relupdown",rel.bin=50)
+  # ),
+  # tar_target(
+  #   name = ls_pred_summary,
+  #   command = multi_summary(ls_pred_models)
+  # ),
+  # tar_target(
+  #   name = ls_pred_mids_summary,
+  #   command = multi_summary(ls_pred_mids_reg)
+  # ),
+  # tar_target(
+  #   name = tbl_preds_log_reg,
+  #   command = pred_log_reg(df_pred_data)
+  # ),
+  # tar_target(
+  #   name = tbl_preds_lin_reg,
+  #   command = pred_lin_reg(df_pred_data)
+  # ),
+  # tar_target(
+  #   name = tbl_preds_lin_imp_reg,
+  #   command = pred_lin_reg(df_pred_mids)
+  # ),
+  # tar_target(
+  #   name = list_pred_clusters,
+  #   command = get_clusters(df_events_complete,rm.out = TRUE)
+  # ),
+  # tar_target(
+  #   name = list_pred_clusters_seq,
+  #   command = get_clusters(df_events_complete,n.cl=4)
+  # )#,
+  # tar_target(
+  #   name = list_df_multi_grouping,
+  #   command = multi_grouping_df_list(df_all_data_formatted,args.list=df_mega_list())
+  # ),
+  # tar_target(
+  #   name = list_multi_group_results,
+  #   command = multi_results_list(list_df_multi_grouping)
+  # ),
+  # tar_target(
+  #   name = list_multi_group_results_direct,
+  #   command = df_all_data_formatted |> multi_grouping_df_list(args.list=df_mega_list()) |> multi_cox_performance_test()
+  # )
+  
+  
+  #,
+  # tar_quarto(
+  #   name = report,
+  #   path = "index.qmd"
+  # ),
+  # tar_target(
+  #   name = pa_change_export_files,
+  #   command = quarto::quarto_render(here::here("doc/pa_change.qmd"), output_format = "docx")
+  # )
+)
+
+## TODO
+## - modify pipeline to only perform imputation once and have the rest use this single object.