fix: a few bugs and update exclusions

Author: dshemetov

R/utils.R

@@ -156,7 +156,6 @@ get_exclusions <- function(
 
 data_substitutions <- function(dataset, disease, forecast_generation_date) {
   disease <- "flu"
-  forecast_generation_date <- as.Date("2025-01-08")
   substitutions <- readr::read_csv(
     glue::glue("{disease}_data_substitutions.csv"),
     comment = "#",

covid_geo_exclusions.csv

@@ -8,6 +8,14 @@ forecast_date,forecaster,geo_value,weight
 "2024-10-01", "climate_base", "all", 2
 "2024-10-01", "climate_geo_agged", "all", 0.5
 "2024-10-01", "climate_quantile_extrapolated", "all", 0
+# feb 12
+2025-02-05, all, mp, 0
+2025-02-05, windowed_seasonal, all, 3
+2025-02-05, windowed_seasonal_extra_sources, all, 0.0
+2025-02-05, linear, all, 0.5
+2025-02-05, linearlog, all, 0
+2025-02-05, climate_base, all, 0
+2025-02-05, climate_geo_agged, all, 0.0
 # feb 5
 2025-02-05, all, mp, 0
 2025-02-05, windowed_seasonal, all, 3

flu_data_substitutions.csv

@@ -1,3 +1,7 @@
 geo_value, forecast_date, time_value, value
 sd, 2025-01-08, 2025-01-01, 33
 id, 2025-01-08, 2025-01-01, 323
+az, 2025-02-12, 2025-02-12, 850
+de, 2025-02-12, 2025-02-12, 120
+id, 2025-02-12, 2025-02-12, 150
+nm, 2025-02-12, 2025-02-12, 200

scripts/covid_hosp_prod.R

@@ -157,7 +157,7 @@ forecast_targets <- tar_map(
       } else {
         train_data <-
           nhsn_latest_data %>%
-          data_substitutions(disease = "covid") %>%
+          data_substitutions(disease = "covid", last(forecast_generation_dates)) %>%
           as_epi_df(as_of = as.Date(forecast_date_int)) %>%
           mutate(time_value = time_value - 3)
       }
@@ -280,7 +280,7 @@ ensemble_targets <- tar_map(
         forecast_full_filtered %>%
           filter(forecaster %in% c("climate_base", "climate_geo_agged")) %>%
           group_by(geo_value, target_end_date, quantile) %>%
-          summarize(value = mean(value, na.rm = TRUE), .groups = "drop") %>%
+          summarize(forecast_date = as.Date(forecast_date_int), value = mean(value, na.rm = TRUE), .groups = "drop") %>%
           ungroup() %>%
           format_flusight(disease = "covid") %>%
           filter(location %nin% c("60", "66", "78")) %>%
@@ -299,7 +299,7 @@ ensemble_targets <- tar_map(
     command = {
       make_submission_csv
       # only validate if we're saving the result to a hub
-      if (submission_directory != "cache") {
+      if (!backtest_mode && submission_directory != "cache") {
         validation <- validate_submission(
           submission_directory,
           file_path = sprintf("CMU-TimeSeries/%s-CMU-TimeSeries.csv", get_forecast_reference_date(forecast_date_int))
@@ -315,7 +315,7 @@ ensemble_targets <- tar_map(
     command = {
       make_climate_submission_csv
       # only validate if we're saving the result to a hub
-      if (submission_directory != "cache" && submit_climatological) {
+      if (!backtest_mode && submission_directory != "cache") {
         validation <- validate_submission(
           submission_directory,
           file_path = sprintf("CMU-climate_baseline/%s-CMU-climate_baseline.csv", get_forecast_reference_date(forecast_date_int))

scripts/flu_hosp_prod.R

@@ -156,7 +156,7 @@ rlang::list2(
         ) %>%
         filter(version == max(version)) %>%
         select(-version) %>%
-        data_substitutions(disease = "flu") %>%
+        data_substitutions(disease = "flu", last(forecast_generation_dates)) %>%
         as_epi_df(other_keys = "source", as_of = Sys.Date())
       most_recent_result
     },
@@ -383,7 +383,7 @@ rlang::list2(
       command = {
         make_submission_csv
         # only validate if we're saving the result to a hub
-        if (submission_directory != "cache") {
+        if (!backtest_mode && submission_directory != "cache") {
           validation <- validate_submission(
             submission_directory,
             file_path = sprintf("CMU-TimeSeries/%s-CMU-TimeSeries.csv", get_forecast_reference_date(forecast_date_int))
@@ -399,7 +399,7 @@ rlang::list2(
       command = {
         make_climate_submission_csv
         # only validate if we're saving the result to a hub
-        if (submission_directory != "cache" && submit_climatological) {
+        if (!backtest_mode && submission_directory != "cache") {
           validation <- validate_submission(
             submission_directory,
             file_path = sprintf("CMU-climate_baseline/%s-CMU-climate_baseline.csv", get_forecast_reference_date(forecast_date_int))