Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

style: minor styling changes #664

Merged
merged 5 commits into from
Oct 2, 2023
Merged

style: minor styling changes #664

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
edbf643
style: styler
dshemetov Sep 30, 2023
f076411
style: line-length style
dshemetov Sep 30, 2023
e897647
lint: add .lintr
dshemetov Sep 30, 2023
87facb2
lint: update lintr
dshemetov Sep 30, 2023
a9afed7
lint: update lintr
dshemetov Sep 30, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
9 changes: 9 additions & 0 deletions R-packages/evalcast/.lintr
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,9 @@
linters: linters_with_defaults(
line_length_linter(120),
cyclocomp_linter = NULL,
object_length_linter(length = 40L)
)
exclusions: list(
"renv",
"venv"
)
122 changes: 81 additions & 41 deletions R-packages/evalcast/R/get_predictions.R
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
#' Get predictions
#'
#'
#' For each of the provided forecast dates, runs a forecaster using the data
#' that would have been available as of that given forecast date. Returns a list
#' of "predictions cards", where each list element corresponds to a different
Expand All @@ -14,13 +14,12 @@
#' probabilities associated with quantile forecasts for that location and
#' ahead. If your forecaster produces point forecasts, then set `quantile=NA`.
#'
#' One argument to `forecaster` must be named `df_list`. It will be
#' populated with the list of historical data returned by a call
#' to COVIDcast. The list will be the same length as the number of rows in
#' the `signals` tibble (see below).
#' The forecaster will also receive a single `forecast_date` as a named argument.
#' Any additional named arguments can be passed via the `forecaster_args`
#' argument below.
#' One argument to `forecaster` must be named `df_list`. It will be populated
#' with the list of historical data returned by a call to COVIDcast. The list
#' will be the same length as the number of rows in the `signals` tibble (see
#' below). The forecaster will also receive a single `forecast_date` as a
#' named argument. Any additional named arguments can be passed via the
#' `forecaster_args` argument below.
#'
#' Thus, the forecaster should have a signature like
#' `forecaster(df_list = data, forecast_data = forecast_date, ...)`
Expand All @@ -29,49 +28,52 @@
#' @template forecast_dates-template
#' @template incidence_period-template
#' @template apply_corrections-template
#' @param response_data_source String indicating the `data_source` of the response.
#' This is used mainly for downstream evaluation. By default, this will be the
#' same as the `data_source` in the first row of the `signals` tibble.
#' @param response_data_source String indicating the `data_source` of the
#' response. This is used mainly for downstream evaluation. By default, this
#' will be the same as the `data_source` in the first row of the `signals`
#' tibble.
#' @param response_data_signal String indicating the `signal` of the response.
#' This is used mainly for downstream evaluation. By default, this will be the
#' same as the `signal` in the first row in the `signals` tibble.
#' @param forecaster_args a list of additional named arguments to be passed
#' to `forecaster()`. A common use case would be to pass the period ahead
#' (e.g. predict 1 day, 2 days, ..., k days ahead). Note that `ahead` is a
#' required component of the forecaster output (see above).
#' @param parallel_execution FALSE (default), TRUE, or a single positive integer. If TRUE,
#' executes each forecast date prediction in parallel on the number of
#' detected cores available minus 1. If FALSE, the code is run on a single
#' core. If integer, runs in parallel on that many cores, clipping to the
#' number of detected cores available if a greater number is requested. Uses
#' [`bettermc::mclapply`] for parallelism if using more than one core.
#' @param parallel_execution FALSE (default), TRUE, or a single positive
#' integer. If TRUE, executes each forecast date prediction in parallel on the
#' number of detected cores available minus 1. If FALSE, the code is run on a
#' single core. If integer, runs in parallel on that many cores, clipping to
#' the number of detected cores available if a greater number is requested.
#' Uses [`bettermc::mclapply`] for parallelism if using more than one core.
#' @param additional_mclapply_args a named list of additional arguments to pass
#' to [`bettermc::mclapply`] (besides `X`, `FUN`, and
#' `mc.cores`.)
#' @param honest_as_of a boolean that, if true, ensures that the forecast_day, end_day,
#' and as_of are all equal when downloading data. Otherwise, as_of is allowed to be freely
#' set (and defaults to current date if not presented).
#' @param honest_as_of a boolean that, if true, ensures that the forecast_day,
#' end_day, and as_of are all equal when downloading data. Otherwise, as_of is
#' allowed to be freely set (and defaults to current date if not presented).
#' @param offline_signal_dir the directory that stores the cached data for each
#' (signal, forecast day) pair. If this is null, no caching is done and the data is
#' downloaded from covidcast.
#' (signal, forecast day) pair. If this is null, no caching is done and the data
#' is downloaded from covidcast.
#'
#' @template predictions_cards-template
#'
#' @examples \dontrun{
#' baby_predictions = get_predictions(
#' baby_predictions <- get_predictions(
#' baseline_forecaster, "baby",
#' tibble::tibble(
#' data_source = "jhu-csse",
#' signal = "deaths_incidence_num",
#' start_day = "2020-08-15",
#' geo_values = "mi",
#' geo_type = "state"),
#' geo_type = "state"
#' ),
#' forecast_dates = "2020-10-01",
#' incidence_period = "epiweek",
#' forecaster_args = list(
#' incidence_period = "epiweek",
#' ahead = 1:4
#' ))
#' )
#' )
#' }
#'
#' @export
Expand All @@ -89,15 +91,40 @@ get_predictions <- function(forecaster,
additional_mclapply_args = list(),
honest_as_of = TRUE,
offline_signal_dir = NULL) {

assert_that(is_tibble(signals), msg = "`signals` should be a tibble.")
assert_that(xor(honest_as_of, "as_of" %in% names(signals)), msg = "`honest_as_of` should be set if and only if `as_of` is not set. Either remove as_of specification or set honest_as_of to FALSE.")
assert_that(
xor(honest_as_of, "as_of" %in% names(signals)),
msg = paste(
"`honest_as_of` should be set if and only if `as_of` is not set. Either",
" remove as_of specification or set honest_as_of to FALSE."
)
)
incidence_period <- match.arg(incidence_period)

if (incidence_period == "epiweek") rlang::warn("incidence_period of 'epiweek' only selects weekly (Saturday) target end dates, actual 'epiweek' signals not supported.", "evalcast::get_predictions")
if ("time_type" %in% names(signals) && any(signals$time_type == "week" | signals$time_type == "epiweek")) rlang::abort("time_type in the signals arg can only be 'day'.", "evalcast::get_predictions")
if ("lag" %in% names(signals)) rlang::warn("lag in the signals arg will be ignored.", "evalcast::get_predictions")
if ("issues" %in% names(signals)) rlang::warn("issues in the signals arg will be ignored.", "evalcast::get_predictions")
if (incidence_period == "epiweek") {
rlang::warn(
paste(
"incidence_period of 'epiweek' only selects weekly (Saturday) target ",
"end dates, actual 'epiweek' signals not supported."
),
"evalcast::get_predictions"
)
}
if (
"time_type" %in% names(signals) &&
any(signals$time_type == "week" | signals$time_type == "epiweek")
) {
rlang::abort(
"time_type in the signals arg can only be 'day'.",
"evalcast::get_predictions"
)
}
if ("lag" %in% names(signals)) {
rlang::warn("lag in the signals arg will be ignored.", "evalcast::get_predictions")
}
if ("issues" %in% names(signals)) {
rlang::warn("issues in the signals arg will be ignored.", "evalcast::get_predictions")
}

if (rlang::is_bool(parallel_execution) && parallel_execution == TRUE) {
num_cores <- max(1L, parallel::detectCores() - 1L)
Expand All @@ -109,11 +136,14 @@ get_predictions <- function(forecaster,
stop("parallel_execution argument is neither boolean nor integer.")
}

assert_that(rlang::is_named2(additional_mclapply_args) &&
all(rlang::names2(additional_mclapply_args) %in% rlang::fn_fmls_names(bettermc::mclapply)))
assert_that(
rlang::is_named2(additional_mclapply_args) &&
all(rlang::names2(additional_mclapply_args) %in% rlang::fn_fmls_names(bettermc::mclapply))
)

get_predictions_single_date_ <- function(forecast_date) {
preds <- do.call(get_predictions_single_date,
preds <- do.call(
get_predictions_single_date,
list(
forecaster = forecaster,
signals = signals,
Expand All @@ -128,7 +158,14 @@ get_predictions <- function(forecaster,
}

if (num_cores > 1L) {
out <- rlang::inject(bettermc::mclapply(forecast_dates, get_predictions_single_date_, mc.cores = num_cores, !!!additional_mclapply_args)) %>% bind_rows()
out <- rlang::inject(
bettermc::mclapply(
forecast_dates,
get_predictions_single_date_,
mc.cores = num_cores,
!!!additional_mclapply_args
)
) %>% bind_rows()
} else {
out <- lapply(forecast_dates, get_predictions_single_date_) %>% bind_rows()
}
Expand All @@ -143,9 +180,10 @@ get_predictions <- function(forecaster,
target_end_date = get_target_period(
.data$forecast_date,
incidence_period,
.data$ahead)$end,
.data$ahead
)$end,
incidence_period = incidence_period
) %>%
) %>%
relocate(.data$forecaster, .before = .data$forecast_date)

class(out) <- c("predictions_cards", class(out))
Expand All @@ -159,7 +197,6 @@ get_predictions_single_date <- function(forecaster,
forecaster_args,
honest_as_of = TRUE,
offline_signal_dir = NULL) {

forecast_date <- lubridate::ymd(forecast_date)
signals <- signal_listcols(signals, forecast_date)

Expand Down Expand Up @@ -189,9 +226,12 @@ get_predictions_single_date <- function(forecaster,

out <- do.call(forecaster, forecaster_args)
assert_that(all(c("ahead", "geo_value", "quantile", "value") %in% names(out)),
msg = paste("Your forecaster must return a data frame with",
"(at least) the columnns `ahead`, `geo_value`,",
"`quantile`, and `value`."))
msg = paste(
"Your forecaster must return a data frame with",
"(at least) the columnns `ahead`, `geo_value`,",
"`quantile`, and `value`."
)
)

# make a predictions card
out$forecast_date <- forecast_date
Expand Down