Merge pull request #201 from cmu-delphi/seasonSummaryPresentation

season summary presentation

Author: dsweber2

.github/workflows/tests.yaml

@@ -38,7 +38,7 @@ jobs:
 
       - name: System dependencies
         run: |
-          sudo apt-get update && sudo apt-get -y install libglpk-dev libcurl4-openssl-dev libfontconfig1-dev libfreetype6-dev libpng-dev libtiff5-dev libjpeg-dev
+          sudo apt-get update && sudo apt-get -y install libglpk-dev libcurl4-openssl-dev libfontconfig1-dev libfreetype6-dev libpng-dev libtiff5-dev libjpeg-dev libharfbuzz-dev libfribidi-dev
 
       - uses: r-lib/actions/setup-renv@v2
 

.gitignore

@@ -23,4 +23,11 @@ data/
 .nhsn_flu_cache.parquet
 meta/
 **/unnamed-chunk*
-decreasing_forecasters_cache/
+decreasing_forecasters_cache/
+reports/tachyons-minimal.css
+reports/season_summary_2025_presentation_files/
+reports/gfx/
+scripts/reports/season_2025_talk/season_summary_2025_presentation_files/
+scripts/reports/season_2025_talk/season_summary_2025_presentation.html
+scripts/reports/season_2025_talk/season_summary_2025_presentation_cache/
+test_proj

.renvignore

@@ -4,6 +4,7 @@ local_logs/
 .github
 .exploration_cache
 .netlify
+.git
 aux_data
 cache/
 covid_hosp_explore

Makefile

@@ -1,22 +1,16 @@
+.PHONY: all test test-forecasters run sync download upload dashboard
+
 current_date:=$(shell date +%F)
 
 install:
 	Rscript -e "install.packages(c('renv', 'pak', 'rspm')); renv::restore()"
 
-.PHONY: all test test-forecasters run run-nohup sync download upload dashboard
-
 test:
 	Rscript -e "testthat::test_dir('tests/testthat')"
 
 run:
 	Rscript scripts/run.R
 
-run-nohup:
-	nohup Rscript scripts/run.R &
-
-run-nohup-restarting:
-	scripts/hardRestarting.sh &
-
 prod-covid:
 	export TAR_RUN_PROJECT=covid_hosp_prod; Rscript scripts/run.R
 
@@ -25,6 +19,14 @@ prod-flu:
 
 prod: prod-covid prod-flu update-site netlify
 
+prod-covid-backtest:
+	export BACKTEST_MODE=TRUE; export TAR_RUN_PROJECT=covid_hosp_prod; Rscript scripts/run.R
+
+prod-flu-backtest:
+	export BACKTEST_MODE=TRUE; export TAR_RUN_PROJECT=flu_hosp_prod; Rscript scripts/run.R
+
+prod-backtest: prod-covid-backtest prod-flu-backtest
+
 explore-covid:
 	export TAR_RUN_PROJECT=covid_hosp_explore; Rscript scripts/run.R
 
@@ -33,6 +35,20 @@ explore-flu:
 
 explore: explore-covid explore-flu update-site netlify
 
+prune: prune-covid-prod prune-flu-prod prune-covid-explore prune-flu-explore
+
+prune-covid-prod:
+	export TAR_PROJECT=covid_hosp_prod; export BACKTEST_MODE=TRUE; Rscript -e "targets::tar_prune()"
+
+prune-flu-prod:
+	export TAR_PROJECT=flu_hosp_prod; export BACKTEST_MODE=TRUE; Rscript -e "targets::tar_prune()"
+
+prune-covid-explore:
+	export TAR_PROJECT=covid_hosp_explore; Rscript -e "targets::tar_prune()"
+
+prune-flu-explore:
+	export TAR_PROJECT=flu_hosp_explore; Rscript -e "targets::tar_prune()"
+
 commit-covid:
 	cd ../covid19-forecast-hub; \
 	git pull --rebase --autostash origin main; \
@@ -50,15 +66,19 @@ commit-flu:
 	git push --force delphi main
 
 submit-covid: commit-covid
-	gh pr create --title "CMU-TimeSeries $(current_date)" --repo cdcgov/covid19-forecast-hub
+	cd ../covid19-forecast-hub; \
+	gh pr create --title "CMU-TimeSeries $(current_date)" --repo CDCgov/covid19-forecast-hub
 
 submit-flu: commit-flu
+	cd ../FluSight-forecast-hub; \
 	gh pr create --title "CMU-TimeSeries $(current_date)" --repo cdcepi/FluSight-forecast-hub
 
 submit-covid-dry: commit-covid
-	gh pr create --title "CMU-TimeSeries $(current_date)" --repo cdcgov/covid19-forecast-hub --dry-run
+	cd ../covid19-forecast-hub; \
+	gh pr create --title "CMU-TimeSeries $(current_date)" --repo CDCgov/covid19-forecast-hub --dry-run
 
 submit-flu-dry: commit-flu
+	cd ../FluSight-forecast-hub; \
 	gh pr create --title "CMU-TimeSeries $(current_date)" --repo cdcepi/FluSight-forecast-hub --dry-run
 
 submit: submit-covid submit-flu
@@ -71,23 +91,49 @@ get-nwss:
 	python nwss_covid_export.py; \
 	python nwss_influenza_export.py
 
-sync:
-	Rscript -e "source('R/sync_aws.R'); sync_aws()"
+pull-aux-data:
+	aws s3 sync s3://forecasting-team-data/2024/aux_data/ aux_data/ --delete
+
+pull-covid-prod:
+	aws s3 sync s3://forecasting-team-data/2024/covid_hosp_prod/ covid_hosp_prod/ --delete
+
+pull-flu-prod:
+	aws s3 sync s3://forecasting-team-data/2024/flu_hosp_prod/ flu_hosp_prod/ --delete
+
+pull-covid-explore:
+	aws s3 sync s3://forecasting-team-data/2024/covid_hosp_explore/ covid_hosp_explore/ --delete
+
+pull-flu-explore:
+	aws s3 sync s3://forecasting-team-data/2024/flu_hosp_explore/ flu_hosp_explore/ --delete
+
+pull: pull-aux-data pull-covid-prod pull-flu-prod pull-covid-explore pull-flu-explore
 
-pull:
-	Rscript -e "source('R/sync_aws.R'); sync_aws(direction = 'download')"
+download: pull
 
-download:pull
+push-covid-prod:
+	aws s3 sync covid_hosp_prod/ s3://forecasting-team-data/2024/covid_hosp_prod/ --delete
 
-push:
-	Rscript -e "source('R/sync_aws.R'); sync_aws(direction = 'upload')"
+push-flu-prod:
+	aws s3 sync flu_hosp_prod/ s3://forecasting-team-data/2024/flu_hosp_prod/ --delete
+
+push-covid-explore:
+	aws s3 sync covid_hosp_explore/ s3://forecasting-team-data/2024/covid_hosp_explore/ --delete
+
+push-flu-explore:
+	aws s3 sync flu_hosp_explore/ s3://forecasting-team-data/2024/flu_hosp_explore/ --delete
+
+push: push-covid-prod push-flu-prod push-covid-explore push-flu-explore
 
 upload: push
 
 dashboard:
 	Rscript scripts/dashboard.R
 
-update-site:
+sync-reports:
+	aws s3 sync s3://forecasting-team-data/2024/reports/ reports/; \
+	aws s3 sync reports/ s3://forecasting-team-data/2024/reports/
+
+update-site: sync-reports
 	Rscript -e "suppressPackageStartupMessages(source(here::here('R', 'load_all.R'))); update_site()" > cache/update_site_log.txt
 
 netlify:
@@ -99,5 +145,16 @@ get-flu-prod-errors:
 get-covid-prod-errors:
 	Rscript -e "suppressPackageStartupMessages(source(here::here('R', 'load_all.R'))); get_targets_errors(project = 'covid_hosp_prod')"
 
-summary_reports:
-	Rscript scripts/summary_reports.R
+summary-reports:
+	Rscript -e "rmarkdown::render('scripts/reports/revision_summary_report_2025.Rmd', output_file = here::here('reports', 'revision_summary_2025.html'))"; \
+	Rscript -e "rmarkdown::render('scripts/reports/decreasing_forecasters.Rmd', output_file = here::here('reports', 'decreasing_forecasters.html'))"; \
+	Rscript -e "rmarkdown::render('scripts/reports/season_summary_2025.Rmd', output_file = here::here('reports', 'season_summary_2025.html'))"; \
+	Rscript -e "rmarkdown::render('scripts/reports/first_day_wrong.Rmd', output_file = here::here('reports', 'first_day_wrong.html'))";
+
+season-summary-2025-talk:
+	quarto render scripts/reports/season_2025_talk/season_summary_2025_presentation.qmd --to revealjs --output-dir "../../../reports"; \
+	cp scripts/reports/season_2025_talk/tachyons-minimal.css reports/; \
+	cp -r scripts/reports/season_2025_talk/gfx reports/
+
+season-summary-2025-talk-preview:
+	quarto preview scripts/reports/season_2025_talk/season_summary_2025_presentation.qmd

NAMESPACE

@@ -280,7 +280,7 @@ drop_non_seasons <- function(epi_data, min_window = 12) {
 
 get_nwss_coarse_data <- function(disease = c("covid", "flu")) {
   disease <- arg_match(disease)
-  aws.s3::get_bucket_df(prefix = glue::glue("exploration/aux_data/nwss_{disease}_data"), bucket = "forecasting-team-data") %>%
+  aws.s3::get_bucket_df(prefix = glue::glue("2024/aux_data/nwss_{disease}_data"), bucket = "forecasting-team-data") %>%
     slice_max(LastModified) %>%
     pull(Key) %>%
     aws.s3::s3read_using(FUN = readr::read_csv, object = ., bucket = "forecasting-team-data")
@@ -312,13 +312,14 @@ add_hhs_region_sum <- function(archive_data_raw, hhs_region_table) {
   archive_data_raw
 }
 
-#' hhs data in covidcast currently
+#' Get versioned NHSN data from healthdata.gov because covidcast API has
+#' incorrect historical data for 2023-2024 season.
 get_health_data <- function(as_of, disease = c("covid", "flu")) {
   as_of <- as.Date(as_of)
   disease <- arg_match(disease)
   checkmate::assert_date(as_of, min.len = 1, max.len = 1)
 
-  cache_path <- here::here("aux_data", "healthdata")
+  cache_path <- here::here("cache", "healthdata")
   if (!dir.exists(cache_path)) {
     dir.create(cache_path, recursive = TRUE)
   }

R/aux_data_utils.R

@@ -353,12 +353,15 @@ compute_pca <- function(epi_data, disease = "flu", ahead = 0, scale_method = "qu
   used_data <- epi_data %>%
     filter(time_value < filter_time) %>%
     select(geo_value, season, source, season_week, hhs)
-  cache_file_path <- paste0(
-    "aux_data/seasonal_features/",
-    disease[grepl("flu|covid", disease)],
-    rlang::hash(used_data),
-    ".parquet",
-    sep = "_"
+  cache_file_path <- here::here(
+    "cache",
+    "seasonal_features",
+    paste0(
+      disease[grepl("flu|covid", disease)],
+      rlang::hash(used_data),
+      ".parquet",
+      sep = "_"
+    )
   )
   if (file.exists(cache_file_path)) {
     seasonal_features <- qs::qread(cache_file_path)

R/forecasters/data_transforms.R

@@ -88,13 +88,13 @@ slide_forecaster <- function(epi_archive,
 epix_slide_simple <- function(epi_archive, forecaster, ref_time_values, before = Inf, cache_key = NULL) {
   # this is so that changing the object without changing the name doesn't result in pulling the wrong cache
   cache_hash <- rlang::hash(epi_archive)
-  dir.create(".exploration_cache/slide_cache", showWarnings = FALSE, recursive = TRUE)
+  dir.create("cache/slide_cache", showWarnings = FALSE, recursive = TRUE)
   out <- purrr::map(ref_time_values, function(tv) {
     if (is.null(cache_key)) {
       epi_df <- epi_archive %>%
         epix_as_of(min(tv, .$versions_end), min_time_value = tv - before)
     } else {
-      file_path <- glue::glue(".exploration_cache/slide_cache/{cache_key}_{cache_hash}_{before}_{tv}.parquet")
+      file_path <- glue::glue("cache/slide_cache/{cache_key}_{cache_hash}_{before}_{tv}.parquet")
       if (file.exists(file_path)) {
         epi_df <- qs::qread(file_path)
       } else {