Skip to contents

Changelog

Source: NEWS.md

epipredict 0.2.5

epipredict 0.2.4

  • Fix flatline_forecaster() to return one prediction per geographic key when the input epi_df has trailing rows with NAs in the outcome (#454). Previously, the forecast was duplicated once per trailing-NA day.

epipredict 0.2.3

  • Fix print.canned_epipred() so the latency-adjustment information actually displays for canned forecasters that include step_adjust_latency in their recipe (#447).

epipredict 0.2.2

epipredict 0.2.1

  • Fix bug in flusight_hub_formatter() so that it works as expected even if the user has not first loaded the epidatasets package.

epipredict 0.2

Breaking changes

  • Moved example datasets from being hosted in the package to being loaded from the epidatasets package. The datasets can no longer be loaded with data(<dataset name>), but can be accessed with data(<dataset name>, package = "epidatasets"), epidatasets::<dataset name> or, after loading the package, the name of the dataset alone (#382).
  • step_adjust_latency() no longer allows empty column selection.
  • Addresses upstream breaking changes from cmu-delphi/epiprocess#595 (growth_rate()). step_growth_rate() has lost its additional_gr_args_list argument and now has an na_rm argument.
  • Moves epiprocess out of depends (#440). No internals have changed, but downstream users may need to add library(epiprocess) to existing code.
  • Removes dependence on the distributional package, replacing the quantiles with hardhat::quantile_pred(). Some associated functions are deprecated with lifecycle messages.
  • Rename check_enough_train_data() to check_enough_data(), and generalize it enough to use as a check on either training or testing.
  • Add check for enough data to predict in arx_forecaster()
  • Adds the .facet_filter option in epiprocess::autoplot() (cmu-delphi/epiprocess#647).

Improvements

  • Add step_adjust_latency, which give several methods to adjust the forecast if the forecast_date is after the last day of data.
  • Fix layer_population_scaling default by with other_keys.
  • Make key column inference more consistent within the package and with current epiprocess.
  • Fix quantile_reg() producing error when asked to output just median-level predictions.
  • (temporary) ahead negative is allowed for step_epi_ahead until we have step_epi_shift
  • Add reference_date as an argument to epi_recipe()
  • Add step_climate() to create “climate” predictor in forecast workflows
  • Add climatological_forecaster() to automatically create climate baselines
  • Replace dist_quantiles() with hardhat::quantile_pred()
  • Allow quantile() to threshold to an interval if desired (#434)
  • arx_forecaster() detects if there’s enough data to predict
  • Add observed_response to autoplot so that forecasts can be plotted against the values they’re predicting

Bug fixes

  • Shifting no columns results in no error for either step_epi_ahead and step_epi_lag
  • Quantiles produced by grf were sometimes out of order.
  • dist_quantiles can have all NA values without causing unrelated errors
  • adjust default quantiles throughout so that they match.
  • force layer_residual_quantiles() to always include 0.5.
  • Rename recipes:::check_training_set() to recipes:::validate_training_data(), as it changed in recipes 1.1.0.
  • A new column name duplicating an existing column name results in an error instead of a random name.

epipredict 0.1

  • simplify layer_residual_quantiles() to avoid timesuck in utils::methods()
  • rename the dist_quantiles() to be more descriptive, breaking change
  • removes previous pivot_quantiles() (now *_wider(), breaking change)
  • add pivot_quantiles_wider() for easier plotting
  • add complement pivot_quantiles_longer()
  • add cdc_baseline_forecaster() and flusight_hub_formatter()
  • add smooth_quantile_reg()
  • improved printing of various methods / internals
  • canned forecasters get a class
  • fixed quantile bug in flatline_forecaster()
  • add functionality to output the unfit workflow from the canned forecasters
  • add quantile_reg()
  • clean up documentation bugs
  • add smooth_quantile_reg()
  • add classifier
  • training window step debugged
  • min_train_window argument removed from canned forecasters
  • add forecasters
  • implement post-processing
  • vignettes avaliable
  • arx_forecaster
  • pkgdown
  • Publish public for easy navigation
  • Two simple forecasters as test beds
  • Working vignette
  • use checkmate for input validation
  • refactor quantile extrapolation (possibly creates different results)
  • force target_date + forecast_date handling to match the time_type of the epi_df. allows for annual and weekly data
  • add check_enough_train_data() that will error if training data is too small
  • added check_enough_train_data() to arx_forecaster()
  • layer_residual_quantiles() will now error if any of the residual quantiles are NA
  • *_args_list() functions now warn if forecast_date + ahead != target_date
  • the predictor argument in arx_forecaster() now defaults to the value of the outcome argument
  • arx_fcast_epi_workflow() and arx_class_epi_workflow() now default to trainer = parsnip::logistic_reg() to match their more canned versions.
  • add a forecast() method simplify generating forecasts
  • refactor bake.epi_recipe() and remove epi_juice().
  • Revise compat-purrr to use the r-lang standalone-* version (via usethis)
  • Replaced old version-faithful example in sliding AR & ARX forecasters vignette
  • epi_recipe() will now warn when given non-epi_df data
  • layer_predict() and predict.epi_workflow() will now appropriately forward ... args intended for predict.model_fit()
  • bake.epi_recipe() will now re-infer the geo and time type in case baking the steps has changed the appropriate values
  • produce length 0 dist_quantiles()
  • add functionality to calculate weighted interval scores for dist_quantiles()
  • Add step_epi_slide to produce generic sliding computations over an epi_df
  • Add quantile random forests (via grf) as a parsnip engine
  • Replace epi_keys() with epiprocess::key_colnames(), #352
  • More descriptive error messages from arg_is_*(), #287
  • Fix bug where fit() drops the epi_workflow class (also error if non-epi_df data is given to epi_recipe()), #363
  • Try to retain the epi_df class during baking to the extent possible, #376