Ndefries/epidatasets migration

DESCRIPTION

@@ -6,13 +6,19 @@ Authors@R: c(
     person("Ryan", "Tibshirani", , "[email protected]", role = "aut"),
     person("Dmitry", "Shemetov", email = "[email protected]", role = "aut"),
     person("David", "Weber", email = "[email protected]", role = "aut"),
-    person("CMU's Delphi Research Group", role = c("cph", "fnd")),
+    person("CMU's Delphi Research Group", role = c("cph", "fnd", "dtc"),
+        comment = "Owner of masking, social-distancing, and CLI data from the COVID-19 Trends and Impacts Survey. Owner of claims-based CLI data from the Delphi Epidata API"),
     person("Logan", "Brooks", role = "aut"),
     person("Rachel", "Lobay", role = "aut"),
     person("Maggie", "Liu", role = "ctb"),
     person("Ken", "Mawer", role = "ctb"),
     person("Chloe", "You", role = "ctb"),
-    person("Jacob", "Bien", role = "ctb")
+    person("Jacob", "Bien", role = "ctb"),
+    person("Johns Hopkins University Center for Systems Science and Engineering", role = "dtc", comment = "Owner of COVID-19 cases and deaths data from the COVID-19 Data Repository"),
+    person("Johns Hopkins University", role = "cph", comment = "Copyright holder of COVID-19 cases and deaths data from the COVID-19 Data Repository"),
+    person("The COVID-19 Canada Open Data Working Group", role = "dtc", comment = "Owner of Canadian COVID-19 cases rates from the Covid19Canada data repository"),
+    person("Statistics Canada", role = "dtc", comment = "Owner of Canadian graduate employment income data from the Statistics Canada website"),
+    person("Google", role = "dtc", comment = "Collaborator on CLI data from the Google symptom surveys")
   )
 Description: A forecasting "framework" for creating epidemiological
     forecasts from versioned data. The framework is designed to be modular
@@ -32,6 +38,7 @@ Imports:
     cli,
     distributional,
     dplyr,
+    epidatasets,
     generics,
     ggplot2,
     glue,
@@ -69,6 +76,7 @@ Suggests:
 VignetteBuilder:
     knitr
 Remotes:
+    cmu-delphi/epidatasets,
     cmu-delphi/epidatr,
     cmu-delphi/epiprocess,
     dajmcdon/smoothqr

NAMESPACE

@@ -139,6 +139,7 @@ export(add_model)
 export(adjust_epi_recipe)
 export(adjust_frosting)
 export(apply_frosting)
+export(archive_cases_dv_subset_all_states)
 export(arx_args_list)
 export(arx_class_args_list)
 export(arx_class_epi_workflow)
@@ -147,10 +148,17 @@ export(arx_fcast_epi_workflow)
 export(arx_forecaster)
 export(autoplot)
 export(bake)
+export(can_prov_cases)
+export(case_death_rate_archive)
+export(cases_deaths_subset)
 export(cdc_baseline_args_list)
 export(cdc_baseline_forecaster)
 export(check_enough_train_data)
 export(clean_f_name)
+export(counts_subset)
+export(county_smoothed_cli_comparison)
+export(covid_case_death_rates)
+export(ctis_covid_behaviours)
 export(default_epi_recipe_blueprint)
 export(detect_layer)
 export(dist_quantiles)
@@ -169,6 +177,7 @@ export(flusight_hub_formatter)
 export(forecast)
 export(frosting)
 export(get_test_data)
+export(grad_employ_subset)
 export(is_epi_recipe)
 export(is_epi_workflow)
 export(is_layer)
@@ -198,6 +207,7 @@ export(remove_frosting)
 export(remove_model)
 export(slather)
 export(smooth_quantile_reg)
+export(state_census)
 export(step_adjust_latency)
 export(step_epi_ahead)
 export(step_epi_lag)
@@ -215,6 +225,7 @@ export(update_model)
 export(validate_layer)
 export(weighted_interval_score)
 import(distributional)
+import(epidatasets)
 import(epiprocess)
 import(parsnip)
 import(recipes)

NEWS.md

@@ -4,12 +4,20 @@ Pre-1.0.0 numbering scheme: 0.x will indicate releases, while 0.0.x will indicat
 
 # epipredict 0.2
 
-## features
+## Breaking changes
+
+- Moved example datasets from being hosted in the package to being reexported
+  from the `epidatasets` package. The datasets can no longer be loaded with
+  `data()` but can be accessed with `epipredict::` or, after loading the package,
+  the name of the dataset alone (#382).
+
+## Improvements
+
 - Add `step_adjust_latency`, which give several methods to adjust the forecast if the `forecast_date` is after the last day of data.
 - (temporary) ahead negative is allowed for `step_epi_ahead` until we have `step_epi_shift`
 
-## bugfixes
-- shifting no columns results in no error for either `step_epi_ahead` and `step_epi_lag`
+## Bug fixes
+- Shifting no columns results in no error for either `step_epi_ahead` and `step_epi_lag`
 
 # epipredict 0.1
 

R/arx_classifier.R

@@ -27,7 +27,7 @@
 #'
 #' @examples
 #' library(dplyr)
-#' jhu <- case_death_rate_subset %>%
+#' jhu <- covid_case_death_rates %>%
 #'   filter(time_value >= as.Date("2021-11-01"))
 #'
 #' out <- arx_classifier(jhu, "death_rate", c("case_rate", "death_rate"))
@@ -104,7 +104,7 @@ arx_classifier <- function(
 #' @seealso [arx_classifier()]
 #' @examples
 #' library(dplyr)
-#' jhu <- case_death_rate_subset %>%
+#' jhu <- covid_case_death_rates %>%
 #'   filter(time_value >= as.Date("2021-11-01"))
 #'
 #' arx_class_epi_workflow(jhu, "death_rate", c("case_rate", "death_rate"))

R/arx_forecaster.R

@@ -25,7 +25,7 @@
 #' @seealso [arx_fcast_epi_workflow()], [arx_args_list()]
 #'
 #' @examples
-#' jhu <- case_death_rate_subset %>%
+#' jhu <- covid_case_death_rates %>%
 #'   dplyr::filter(time_value >= as.Date("2021-12-01"))
 #'
 #' out <- arx_forecaster(
@@ -96,7 +96,7 @@ arx_forecaster <- function(
 #'
 #' @examples
 #' library(dplyr)
-#' jhu <- case_death_rate_subset %>%
+#' jhu <- covid_case_death_rates %>%
 #'   filter(time_value >= as.Date("2021-12-01"))
 #'
 #' arx_fcast_epi_workflow(

R/autoplot.R

@@ -29,7 +29,7 @@ ggplot2::autoplot
 #' @name autoplot-epipred
 #' @examples
 #' library(dplyr)
-#' jhu <- case_death_rate_subset %>%
+#' jhu <- covid_case_death_rates %>%
 #'   filter(time_value >= as.Date("2021-11-01"))
 #'
 #' r <- epi_recipe(jhu) %>%
@@ -70,7 +70,7 @@ ggplot2::autoplot
 #'
 #' # ------- Plotting canned forecaster output
 #'
-#' jhu <- case_death_rate_subset %>%
+#' jhu <- covid_case_death_rates %>%
 #'   filter(time_value >= as.Date("2021-11-01"))
 #' flat <- flatline_forecaster(jhu, "death_rate")
 #' autoplot(flat, .max_facets = 4)

R/cdc_baseline_forecaster.R

@@ -23,7 +23,7 @@
 #'
 #' @examples
 #' library(dplyr)
-#' weekly_deaths <- case_death_rate_subset %>%
+#' weekly_deaths <- covid_case_death_rates %>%
 #'   select(geo_value, time_value, death_rate) %>%
 #'   left_join(state_census %>% select(pop, abbr), by = c("geo_value" = "abbr")) %>%
 #'   mutate(deaths = pmax(death_rate / 1e5 * pop * 7, 0)) %>%

R/epi_recipe.R

@@ -43,7 +43,7 @@ epi_recipe.default <- function(x, ...) {
 #' @examples
 #' library(dplyr)
 #' library(recipes)
-#' jhu <- case_death_rate_subset %>%
+#' jhu <- covid_case_death_rates %>%
 #'   filter(time_value > "2021-08-01") %>%
 #'   arrange(geo_value, time_value)
 #'
@@ -263,7 +263,7 @@ is_epi_recipe <- function(x) {
 #' library(dplyr)
 #' library(recipes)
 #'
-#' jhu <- case_death_rate_subset %>%
+#' jhu <- covid_case_death_rates %>%
 #'   filter(time_value > "2021-08-01") %>%
 #'   arrange(geo_value, time_value)
 #'
@@ -347,7 +347,7 @@ update_epi_recipe <- function(x, recipe, ..., blueprint = default_epi_recipe_blu
 #' library(dplyr)
 #' library(workflows)
 #'
-#' jhu <- case_death_rate_subset %>%
+#' jhu <- covid_case_death_rates %>%
 #'   filter(time_value > "2021-11-01", geo_value %in% c("ak", "ca", "ny"))
 #' r <- epi_recipe(jhu) %>%
 #'   step_epi_lag(death_rate, lag = c(0, 7, 14)) %>%

R/epi_workflow.R

@@ -20,7 +20,7 @@
 #' @importFrom generics augment
 #' @export
 #' @examples
-#' jhu <- case_death_rate_subset
+#' jhu <- covid_case_death_rates
 #'
 #' r <- epi_recipe(jhu) %>%
 #'   step_epi_lag(death_rate, lag = c(0, 7, 14)) %>%
@@ -84,7 +84,7 @@ is_epi_workflow <- function(x) {
 #' @name fit-epi_workflow
 #' @export
 #' @examples
-#' jhu <- case_death_rate_subset %>%
+#' jhu <- covid_case_death_rates %>%
 #'   filter(time_value > "2021-11-01", geo_value %in% c("ak", "ca", "ny"))
 #'
 #' r <- epi_recipe(jhu) %>%
@@ -142,7 +142,7 @@ fit.epi_workflow <- function(object, data, ..., control = workflows::control_wor
 #' @name predict-epi_workflow
 #' @export
 #' @examples
-#' jhu <- case_death_rate_subset
+#' jhu <- covid_case_death_rates
 #'
 #' r <- epi_recipe(jhu) %>%
 #'   step_epi_lag(death_rate, lag = c(0, 7, 14)) %>%

R/epipredict-package.R

@@ -1,5 +1,5 @@
 ## usethis namespace: start
-#' @import epiprocess parsnip
+#' @import epiprocess parsnip epidatasets
 #' @importFrom checkmate assert_class assert_numeric
 #' @importFrom checkmate test_character test_date test_function
 #' @importFrom checkmate test_integerish test_logical

R/flatline_forecaster.R

@@ -24,7 +24,7 @@
 #' @export
 #'
 #' @examples
-#' jhu <- case_death_rate_subset %>%
+#' jhu <- covid_case_death_rates %>%
 #'   dplyr::filter(time_value >= as.Date("2021-12-01"))
 #'
 #' out <- flatline_forecaster(jhu, "death_rate")

R/flusight_hub_formatter.R

@@ -1,7 +1,6 @@
 location_to_abbr <- function(location) {
   dictionary <-
     state_census %>%
-    mutate(fips = sprintf("%02d", fips)) %>%
     dplyr::transmute(
       location = dplyr::case_match(fips, "00" ~ "US", .default = fips),
       abbr
@@ -12,7 +11,6 @@ location_to_abbr <- function(location) {
 abbr_to_location <- function(abbr) {
   dictionary <-
     state_census %>%
-    mutate(fips = sprintf("%02d", fips)) %>%
     dplyr::transmute(
       location = dplyr::case_match(fips, "00" ~ "US", .default = fips),
       abbr
@@ -57,7 +55,7 @@ abbr_to_location <- function(abbr) {
 #'
 #' @examples
 #' library(dplyr)
-#' weekly_deaths <- case_death_rate_subset %>%
+#' weekly_deaths <- covid_case_death_rates %>%
 #'   filter(
 #'     time_value >= as.Date("2021-09-01"),
 #'     geo_value %in% c("ca", "ny", "dc", "ga", "vt")

R/frosting.R

@@ -9,7 +9,7 @@
 #'
 #' @examples
 #' library(dplyr)
-#' jhu <- case_death_rate_subset %>%
+#' jhu <- covid_case_death_rates %>%
 #'   filter(time_value > "2021-11-01", geo_value %in% c("ak", "ca", "ny"))
 #' r <- epi_recipe(jhu) %>%
 #'   step_epi_lag(death_rate, lag = c(0, 7, 14)) %>%
@@ -128,7 +128,7 @@ update_frosting <- function(x, frosting, ...) {
 #' @export
 #' @examples
 #' library(dplyr)
-#' jhu <- case_death_rate_subset %>%
+#' jhu <- covid_case_death_rates %>%
 #'   filter(time_value > "2021-11-01", geo_value %in% c("ak", "ca", "ny"))
 #' r <- epi_recipe(jhu) %>%
 #'   step_epi_lag(death_rate, lag = c(0, 7, 14)) %>%
@@ -268,7 +268,7 @@ new_frosting <- function() {
 #' wf <- epi_workflow() %>% add_frosting(f)
 #'
 #' # A more realistic example
-#' jhu <- case_death_rate_subset %>%
+#' jhu <- covid_case_death_rates %>%
 #'   filter(time_value > "2021-11-01", geo_value %in% c("ak", "ca", "ny"))
 #'
 #' r <- epi_recipe(jhu) %>%

R/get_test_data.R

@@ -19,11 +19,11 @@
 #'   keys, as well other variables in the original dataset.
 #' @examples
 #' # create recipe
-#' rec <- epi_recipe(case_death_rate_subset) %>%
+#' rec <- epi_recipe(covid_case_death_rates) %>%
 #'   step_epi_ahead(death_rate, ahead = 7) %>%
 #'   step_epi_lag(death_rate, lag = c(0, 7, 14)) %>%
 #'   step_epi_lag(case_rate, lag = c(0, 7, 14))
-#' get_test_data(recipe = rec, x = case_death_rate_subset)
+#' get_test_data(recipe = rec, x = covid_case_death_rates)
 #' @importFrom rlang %@%
 #' @importFrom stats na.omit
 #' @export