Merge pull request #71 from diazrenata/diss

Include 95th percentile of sim dissimilarity scores (directly shows how dissimilar observed would need to be to be "Extreme")

Author: diazrenata

R/metrics.R

@@ -491,6 +491,7 @@ po_central_tendency <- function(fs_df, fs_po_df) {
       real_r2 = real_rsq,
       best_po_sim = ct_sim,
       sim_pos_from_best = mean(ct_pos$prop_off),
+      sim_pos_from_best_95 = quantile(ct_pos$prop_off, probs = .95, na.rm = T),
       sim_r2_from_best = mean(ct_pos$rsq),
       sim_r2_from_best_median = median(ct_pos$rsq),
       ncomparisons = nrow(ct_pos),

analysis/bbs_pipeline.R

@@ -53,27 +53,27 @@ cache$del(key = "lock", namespace = "session")
 
 ## Run the pipeline
 nodename <- Sys.info()["nodename"]
-if(grepl("ufhpc", nodename)) {
-  print("I know I am on the HiPerGator!")
-  library(clustermq)
-  options(clustermq.scheduler = "slurm", clustermq.template = here::here("slurm_clustermq.tmpl"))
-  ## Run the pipeline parallelized for HiPerGator
-  make(all,
-       force = TRUE,
-       cache = cache,
-       cache_log_file = here::here("analysis", "drake", "cache_log_bbs.txt"),
-       verbose = 1,
-       parallelism = "clustermq",
-       jobs = 100,
-       caching = "master",
-       memory_strategy = "autoclean",
-       garbage_collection = TRUE) # Important for DBI caches!
-} else {
+# if(grepl("ufhpc", nodename)) {
+#   print("I know I am on the HiPerGator!")
+#   library(clustermq)
+#   options(clustermq.scheduler = "slurm", clustermq.template = here::here("slurm_clustermq.tmpl"))
+#   ## Run the pipeline parallelized for HiPerGator
+#   make(all,
+#        force = TRUE,
+#        cache = cache,
+#        cache_log_file = here::here("analysis", "drake", "cache_log_bbs.txt"),
+#        verbose = 1,
+#        parallelism = "clustermq",
+#        jobs = 100,
+#        caching = "master",
+#        memory_strategy = "autoclean",
+#        garbage_collection = TRUE) # Important for DBI caches!
+# } else {
   library(clustermq)
   options(clustermq.scheduler = "multicore")
   # Run the pipeline on multiple local cores
   system.time(make(all, cache = cache, cache_log_file = here::here("analysis", "drake", "cache_log_bbs.txt"), verbose = 1, memory_strategy = "autoclean"))
-}
+#}
 
 #system.time(make(all, cache = cache, cache_log_file = here::here("analysis", "drake", "cache_log_bbs.txt")))
 

analysis/gentry_pipeline.R

@@ -51,27 +51,27 @@ cache$del(key = "lock", namespace = "session")
 
 ## Run the pipeline
 nodename <- Sys.info()["nodename"]
-if(grepl("ufhpc", nodename)) {
-  print("I know I am on the HiPerGator!")
-  library(clustermq)
-  options(clustermq.scheduler = "slurm", clustermq.template = here::here("slurm_clustermq.tmpl"))
-  ## Run the pipeline parallelized for HiPerGator
-  make(all,
-       force = TRUE,
-       cache = cache,
-       cache_log_file = here::here("analysis", "drake", "cache_log_gentry.txt"),
-       verbose = 1,
-       parallelism = "clustermq",
-       jobs = 100,
-       caching = "master",
-       memory_strategy = "autoclean",
-       garbage_collection = TRUE) # Important for DBI caches!
-} else {
+# if(grepl("ufhpc", nodename)) {
+#   print("I know I am on the HiPerGator!")
+#   library(clustermq)
+#   options(clustermq.scheduler = "slurm", clustermq.template = here::here("slurm_clustermq.tmpl"))
+#   ## Run the pipeline parallelized for HiPerGator
+#   make(all,
+#        force = TRUE,
+#        cache = cache,
+#        cache_log_file = here::here("analysis", "drake", "cache_log_gentry.txt"),
+#        verbose = 1,
+#        parallelism = "clustermq",
+#        jobs = 100,
+#        caching = "master",
+#        memory_strategy = "autoclean",
+#        garbage_collection = TRUE) # Important for DBI caches!
+# } else {
   library(clustermq)
   options(clustermq.scheduler = "multicore")
   # Run the pipeline on multiple local cores
   system.time(make(all, cache = cache, cache_log_file = here::here("analysis", "drake", "cache_log_gentry.txt"), verbose = 1, memory_strategy = "autoclean"))
-}
+#}
 
 #system.time(make(all, cache = cache, cache_log_file = here::here("analysis", "drake", "cache_log_gentry.txt")))
 

analysis/manuscript/finaledit/Appendices/ELEDiazGraphicalAbstract.pdf

@@ -48,30 +48,30 @@ all <- drake_plan(
 db <- DBI::dbConnect(RSQLite::SQLite(), here::here("analysis", "drake", "drake-cache-mcdb.sqlite"))
 cache <- storr::storr_dbi("datatable", "keystable", db)
 cache$del(key = "lock", namespace = "session")
-
-## Run the pipeline
-nodename <- Sys.info()["nodename"]
-if(grepl("ufhpc", nodename)) {
- print("I know I am on the HiPerGator!")
- library(clustermq)
- options(clustermq.scheduler = "slurm", clustermq.template = here::here("slurm_clustermq.tmpl"))
- ## Run the pipeline parallelized for HiPerGator
- make(all,
-      force = TRUE,
-      cache = cache,
-      cache_log_file = here::here("analysis", "drake", "cache_log_mcdb.txt"),
-      verbose = 1,
-      parallelism = "clustermq",
-      jobs = 20,
-      caching = "master",
-     memory_strategy = "autoclean",
-     garbage_collection = TRUE) # Important for DBI caches!
-} else {
+#
+# ## Run the pipeline
+# nodename <- Sys.info()["nodename"]
+# if(grepl("ufhpc", nodename)) {
+#  print("I know I am on the HiPerGator!")
+#  library(clustermq)
+#  options(clustermq.scheduler = "slurm", clustermq.template = here::here("slurm_clustermq.tmpl"))
+#  ## Run the pipeline parallelized for HiPerGator
+#  make(all,
+#       force = TRUE,
+#       cache = cache,
+#       cache_log_file = here::here("analysis", "drake", "cache_log_mcdb.txt"),
+#       verbose = 1,
+#       parallelism = "clustermq",
+#       jobs = 20,
+#       caching = "master",
+#      memory_strategy = "autoclean",
+#      garbage_collection = TRUE) # Important for DBI caches!
+# } else {
   library(clustermq)
   options(clustermq.scheduler = "multicore")
   # Run the pipeline on multiple local cores
   system.time(make(all, cache = cache, cache_log_file = here::here("analysis", "drake", "cache_log_mcdb.txt"), verbose = 1, memory_strategy = "autoclean"))
-}
+#}
 
 DBI::dbDisconnect(db)
 rm(cache)

analysis/manuscript/finaledit/Appendices/ELEDiazSA1.docx

@@ -53,27 +53,27 @@ cache$del(key = "lock", namespace = "session")
 
 # Run the pipeline
 nodename <- Sys.info()["nodename"]
-if(grepl("ufhpc", nodename)) {
-  print("I know I am on the HiPerGator!")
-  library(clustermq)
-  options(clustermq.scheduler = "slurm", clustermq.template = here::here("slurm_clustermq.tmpl"))
-  ## Run the pipeline parallelized for HiPerGator
-  make(all,
-       force = TRUE,
-       cache = cache,
-       cache_log_file = here::here("analysis", "drake", "cache_log_miscabund.txt"),
-       verbose = 1,
-       parallelism = "clustermq",
-       jobs = 100,
-       caching = "master",
-       memory_strategy = "autoclean",
-       garbage_collection = TRUE) # Important for DBI caches!
-} else {
+# if(grepl("ufhpc", nodename)) {
+#   print("I know I am on the HiPerGator!")
+#   library(clustermq)
+#   options(clustermq.scheduler = "slurm", clustermq.template = here::here("slurm_clustermq.tmpl"))
+#   ## Run the pipeline parallelized for HiPerGator
+#   make(all,
+#        force = TRUE,
+#        cache = cache,
+#        cache_log_file = here::here("analysis", "drake", "cache_log_miscabund.txt"),
+#        verbose = 1,
+#        parallelism = "clustermq",
+#        jobs = 100,
+#        caching = "master",
+#        memory_strategy = "autoclean",
+#        garbage_collection = TRUE) # Important for DBI caches!
+# } else {
   library(clustermq)
   options(clustermq.scheduler = "multicore")
   # Run the pipeline on multiple local cores
   system.time(make(all, cache = cache, cache_log_file = here::here("analysis", "drake", "cache_log_miscabund.txt"), verbose = 1, memory_strategy = "autoclean"))
-}
+#}
 
 #system.time(make(all, cache = cache, cache_log_file = here::here("analysis", "drake", "cache_log_miscabund.txt")))