diff --git a/docs/reference/transmute.harp_fcst.html b/docs/reference/transmute.harp_fcst.html
index 6028722..99a2cf2 100644
--- a/docs/reference/transmute.harp_fcst.html
+++ b/docs/reference/transmute.harp_fcst.html
@@ -24,6 +24,9 @@
diff --git a/docs/search.json b/docs/search.json
index 5e6a633..8fe28ba 100644
--- a/docs/search.json
+++ b/docs/search.json
@@ -1 +1 @@
-[{"path":"/LICENSE.html","id":null,"dir":"","previous_headings":"","what":"MIT License","title":"MIT License","text":"Copyright (c) 2018 Andrew Singleton Permission hereby granted, free charge, person obtaining copy software associated documentation files (“Software”), deal Software without restriction, including without limitation rights use, copy, modify, merge, publish, distribute, sublicense, /sell copies Software, permit persons Software furnished , subject following conditions: copyright notice permission notice shall included copies substantial portions Software. SOFTWARE PROVIDED “”, WITHOUT WARRANTY KIND, EXPRESS IMPLIED, INCLUDING LIMITED WARRANTIES MERCHANTABILITY, FITNESS PARTICULAR PURPOSE NONINFRINGEMENT. EVENT SHALL AUTHORS COPYRIGHT HOLDERS LIABLE CLAIM, DAMAGES LIABILITY, WHETHER ACTION CONTRACT, TORT OTHERWISE, ARISING , CONNECTION SOFTWARE USE DEALINGS SOFTWARE.","code":""},{"path":"/authors.html","id":null,"dir":"","previous_headings":"","what":"Authors","title":"Authors and Citation","text":"Andrew Singleton. Author, maintainer. Alex Deckmyn. Author.","code":""},{"path":"/authors.html","id":"citation","dir":"","previous_headings":"","what":"Citation","title":"Authors and Citation","text":"Singleton , Deckmyn (2023). harpPoint: Point verifition NWP forecasts. R package version 0.2.0, https://github.com/harphub/harpPoint.","code":"@Manual{, title = {harpPoint: Point verifition for NWP forecasts}, author = {Andrew Singleton and Alex Deckmyn}, year = {2023}, note = {R package version 0.2.0}, url = {https://github.com/harphub/harpPoint}, }"},{"path":"/index.html","id":"harppoint-","dir":"","previous_headings":"","what":"Point verifition for NWP forecasts","title":"Point verifition for NWP forecasts","text":"harpPoint provides functionality verification meteorological data geographic points. Typically verification forecasts interpolated locations weather stations. Functions provided computing verification scores deterministic ensemble forecasts. addition, confidence intervals scores, differences scores different forecast models can computed using bootstrapping.","code":""},{"path":"/index.html","id":"installation","dir":"","previous_headings":"","what":"Installation","title":"Point verifition for NWP forecasts","text":"can install harpPoint GitHub :","code":"# install.packages(\"remotes\") remotes::install_github(\"harphub/harpPoint\")"},{"path":"/index.html","id":"verification","dir":"","previous_headings":"","what":"Verification","title":"Point verifition for NWP forecasts","text":"harpPoint functions verification designed work data read using functions harpIO. means harp_df data frames harp_lists. data must include column observations forecasts verified. two main functions verification: det_verify() - deterministic forecasts; ens_verify() - ensemble forecasts. functions output harp_verif list. list data frames scores separated summary scores threshold scores. Threshold scores computed thresholds provided functions computed probabilities threshold exceedance.","code":""},{"path":"/index.html","id":"deterministic-scores","dir":"","previous_headings":"","what":"Deterministic scores","title":"Point verifition for NWP forecasts","text":"det_verify() computes scores following column names:","code":""},{"path":"/index.html","id":"summary-scores","dir":"","previous_headings":"Deterministic scores","what":"Summary scores","title":"Point verifition for NWP forecasts","text":"bias - mean difference forecasts observations rmse - root mean squared error mae - mean absolute error stde - standard deviation error hexbin - heat map paired hexagonal bins forecasts observations","code":""},{"path":"/index.html","id":"threshold-scores","dir":"","previous_headings":"Deterministic scores","what":"Threshold scores","title":"Point verifition for NWP forecasts","text":"cont_tab - contingency table forecast hits, misses, false alarms correct rejections threat_score - ratio hits sum hits, misses false alarms hit_rate - ratio hits sum hits misses miss_rate - ratio misses sum hits misses false_alarm_rate - ratio false alarms sum false alarms correct rejections false_alarm_ratio - ratio false alarms sum false alarms hits heidke_skill_score - fraction correct forecasts eliminating forecasts correct purely due random chance pierce_skill_score - 1 - miss rate - false alarm rate kuiper_skill_score - well forecasts separates hits false alarms percent_correct - ratio sum hits correct rejections total number cases frequency_bias - ratio sum hits false alarms sum hits misses equitable_threat_score - well forecast measures hits accounting hits due pure chance odds_ratio - ratio product hits correct rejections product misses false alarms log_odds_ratio - sum logs hits correct rejections minus sum logs misses false alarms. odds_ratio_skill_score - ratio product hits correct rejections minus product misses false alarms product hits correct rejections plus product misses false alarms extreme_dependency_score - ratio difference logs observations climatology hit rate sum logs observations climatology hit rate symmetric_eds - symmetric extreme dependency score, ratio difference logs forecast climatology hit rate sum logs forecast climatology hit rate extreme_dependency_index - ratio difference logs false alarm rate hit rate sum logs false alarm rate hit rate symmetric_edi - symmetric extreme dependency index, ratio sum difference logs false alarm rate hit rate difference logs inverse hit rate false alarm rate sum logs hit rate, false alarm rate, inverse hit rate inverse false alarm rate. inverse 1 - value.","code":""},{"path":"/index.html","id":"ensemble-scores","dir":"","previous_headings":"","what":"Ensemble scores","title":"Point verifition for NWP forecasts","text":"ens_verify() computes scores following column names:","code":""},{"path":"/index.html","id":"summary-scores-1","dir":"","previous_headings":"Ensemble scores","what":"Summary scores","title":"Point verifition for NWP forecasts","text":"mean_bias - mean difference ensemble mean forecasts observations rmse - root mean squared error stde - standard deviation error spread - square root mean variance ensemble forecasts hexbin - heat map paired hexagonal bins forecasts observations rank_histogram - Observation counts ranked ensemble member bins crps- cumulative rank probability score - difference cumulative distribution ensemble forecasts step function observations crps_potential - crps achieved perfectly reliable ensemble crps_reliability - Measures ability ensemble produce cumulative distribution desired statisical properties. fair_crps - crps achieved either ensemble infinite number members, number members provided function","code":""},{"path":"/index.html","id":"threshold-scores-1","dir":"","previous_headings":"Ensemble scores","what":"Threshold scores","title":"Point verifition for NWP forecasts","text":"brier_score - mean squared error ensemble probability space fair_brier_score - Brier score achieved either ensemble infinite number members, number members provided function brier_skill_score - Brier score compared reference probabilistic forecast (usually observed climatology) brier_score_reliability - measure ensemble’s ability produce reliable (forecast probability = observed frequency) forecasts brier_score_resolution - measure ensemble’s ability discriminate “day” uncertainty climatological uncertainty brier_score_uncertainty - inherent uncertainty events reliability - frequency observations bins forecast probability roc - relative operating characteristic ensemble - hit rates false alarm rates forecast probability bins roc_area - area roc curve - summarises ability ensemble discriminate events non events economic_value - relative improvement economic value forecast compared climatology range cost / loss ratios","code":""},{"path":"/index.html","id":"getting-gridded-data-to-points","dir":"","previous_headings":"","what":"Getting gridded data to points","title":"Point verifition for NWP forecasts","text":"interpolation gridded data points see Interpolate section Transforming model data article harpIO website, documentation geo_points.","code":""},{"path":"/reference/arrange.harp_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Arrange columns from tables in a harp_fcst object. — arrange.harp_fcst","title":"Arrange columns from tables in a harp_fcst object. — arrange.harp_fcst","text":"Works way arrange, except runs tables harp_fcst object. means common columns objects can safely arranged.","code":""},{"path":"/reference/arrange.harp_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Arrange columns from tables in a harp_fcst object. — arrange.harp_fcst","text":"","code":"# S3 method for harp_fcst arrange(.fcst, ...)"},{"path":"/reference/arrange.harp_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Arrange columns from tables in a harp_fcst object. — arrange.harp_fcst","text":".fcst harp_fcst object ... Arguments arrange","code":""},{"path":"/reference/bin_fcst_obs.html","id":null,"dir":"Reference","previous_headings":"","what":"Create a bi-variate histogram of forecast, observation pairs — bin_fcst_obs","title":"Create a bi-variate histogram of forecast, observation pairs — bin_fcst_obs","text":"Values forecasts observations binned bands whereby density forecast, observation pairs bin calculated. hood, data binned hexagons using hexbin. Hexagons used since symmetry nearest neighbours unlike square bins, plot time polygon maximum number sides tessellate.","code":""},{"path":"/reference/bin_fcst_obs.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Create a bi-variate histogram of forecast, observation pairs — bin_fcst_obs","text":"","code":"bin_fcst_obs( .fcst, parameter, groupings = \"lead_time\", num_bins = 30, show_progress = TRUE, ... ) # S3 method for harp_det_point_df bin_fcst_obs( .fcst, parameter, groupings = \"lead_time\", num_bins = 30, show_progress = TRUE, fcst_model = NULL, ... ) # S3 method for harp_ens_point_df bin_fcst_obs( .fcst, parameter, groupings = \"lead_time\", num_bins = 30, show_progress = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/bin_fcst_obs.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Create a bi-variate histogram of forecast, observation pairs — bin_fcst_obs","text":".fcst harp_df data frame harp_list. parameter column containing parameter. Can unquoted, quoted string, embraced variable name (.e var). groupings groupings compute binned densities. Must vector strings, list vectors strings. num_bins number bins partition observations. show_progress Logical. Whether show progress bar. ... Arguments methods. fcst_model name forecast model. .fcst contain fcst_model, new column created populated value. fcst_model column exists, value column replaced value.","code":""},{"path":"/reference/bin_fcst_obs.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Create a bi-variate histogram of forecast, observation pairs — bin_fcst_obs","text":"harp_verif list.","code":""},{"path":[]},{"path":"/reference/bind_bootstrap_score.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Bind output of bootstrap_score / pooled_bootstrap_score — bind_bootstrap_score","text":"","code":"bind_bootstrap_score(...)"},{"path":"/reference/bind_bootstrap_score.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Bind output of bootstrap_score / pooled_bootstrap_score — bind_bootstrap_score","text":"... list outputs bootstrap_score() / pooled_bootstrap_score()","code":""},{"path":"/reference/bind_bootstrap_score.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Bind output of bootstrap_score / pooled_bootstrap_score — bind_bootstrap_score","text":"object class harp_bootstrap","code":""},{"path":"/reference/bind_point_verif.html","id":null,"dir":"Reference","previous_headings":"","what":"Bind harp verification objects into a single object — bind_point_verif","title":"Bind harp verification objects into a single object — bind_point_verif","text":"plotting may desirable combine various different verifications single object. example, verification different parameters set forecast models.","code":""},{"path":"/reference/bind_point_verif.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Bind harp verification objects into a single object — bind_point_verif","text":"","code":"bind_point_verif(...)"},{"path":"/reference/bind_point_verif.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Bind harp verification objects into a single object — bind_point_verif","text":"... Verification objects combined. Can individual objects list.","code":""},{"path":"/reference/bind_point_verif.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Bind harp verification objects into a single object — bind_point_verif","text":"harp verification object attributes moved columns combined new attributes.","code":""},{"path":"/reference/bootstrap_score.html","id":null,"dir":"Reference","previous_headings":"","what":"Bootstrap a score — bootstrap_score","title":"Bootstrap a score — bootstrap_score","text":"Compute confidence intervals verification scores using bootstrap method. one forecast confidence differences forecasts also computed.","code":""},{"path":"/reference/bootstrap_score.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Bootstrap a score — bootstrap_score","text":"","code":"bootstrap_score( .fcst, score_function, parameter, n, groupings = \"leadtime\", confidence_interval = 0.95, ... )"},{"path":"/reference/bootstrap_score.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Bootstrap a score — bootstrap_score","text":".fcst harp_fcst object. score_function name verification function bootstrap . parameter parameter gives name observations columns. n Th number bootstrap replicants. groupings groups compute verification scores . default leadtime. ... arguments score function. e.g. thresholds.","code":""},{"path":"/reference/bootstrap_verify.html","id":null,"dir":"Reference","previous_headings":"","what":"Bootstrap a verification function — bootstrap_verify","title":"Bootstrap a verification function — bootstrap_verify","text":"bootstrap_verify used compute verification scores confidence intervals. one fcst_model exists input harp_list object, statistical significance differences verification scores fcst_models computed. statistical testing done using bootstrap method whereby scores computed repeatedly random samples input data.","code":""},{"path":"/reference/bootstrap_verify.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Bootstrap a verification function — bootstrap_verify","text":"","code":"bootstrap_verify( .fcst, verif_func, obs_col, n, groupings = \"lead_time\", pool_by = NULL, conf = 0.95, min_cases = 4, perfect_scores = perfect_score(), parallel = FALSE, num_cores = NULL, show_progress = TRUE, ... )"},{"path":"/reference/bootstrap_verify.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Bootstrap a verification function — bootstrap_verify","text":".fcst harp_list object column observations. verif_func harpPoint verification function bootstrap. obs_col observations column harp_list object. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}} n number bootstrap replicates. groupings groups compute scores. See group_by information grouping works. pool_by block bootstrap, quoted column name use pool data blocks. overlapping blocks data frame column common harp_list object input column named \"pool\" pool data belong . See Details. conf confidence interval compute. min_cases minimum number cases required group. block bootstrapping minimum number blocks. perfect_scores values score perfect score. parallel Set TRUE use parallel processing bootstrapping. Requires parallel package. num_cores parallel = TRUE, number cores use parallel processing. NULL, number cores detected parallel::detectCores() used. show_progress Logical. Set TRUE show progress bar. feature available parallel = TRUE ... arguments verif_func","code":""},{"path":"/reference/bootstrap_verify.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Bootstrap a verification function — bootstrap_verify","text":"harp_point_verif object extra columns upper lower confidence bounds scores percent replicates \"better\" one fcst_models input harp_list_object","code":""},{"path":"/reference/bootstrap_verify.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Bootstrap a verification function — bootstrap_verify","text":"data auto-correlated block bootstrap may used, whereby data pooled groups serial dependencies maintained. Rather sampling individual data points randomly, pools data points sampled randomly. pools taken column passed pool_by argument. use overlapping block bootstrap data frame passed pool_by, one column common harp_list object input column named \"pool\" labels pool row . ensures correct number overlapping pools used bootstrap replicate. make_bootstrap_pools can used get data frame overlapping pools. Bootstrapping can quite slow since many replicates computed. order speed process , bootstrap_verify also works parallel whereby replicates computed individual cores parallel rather serial. can achieved setting parallel = TRUE. default behaviour use cores, number cores can set num_cores argument.","code":""},{"path":"/reference/check_obs_against_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Observation error check against forecast — check_obs_against_fcst","title":"Observation error check against forecast — check_obs_against_fcst","text":"stratification standard deviation forecast forecast cycles models (mname), case eps forecast, ensemble members computed. difference observation forecast expected smaller number multiples standard deviation. number multiples standard deviation can supplied default value used depending parameter.","code":""},{"path":"/reference/check_obs_against_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Observation error check against forecast — check_obs_against_fcst","text":"","code":"check_obs_against_fcst( .fcst, parameter, num_sd_allowed = NULL, stratification = c(\"SID\", \"quarter_day\") )"},{"path":"/reference/check_obs_against_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Observation error check against forecast — check_obs_against_fcst","text":".fcst harp_df data frame, harp_list, observations column. parameter observations column. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. num_sd_allowed number standard deviations forecast difference forecast observation must smaller . stratification columns stratify data computing allowed tolerance. cases column must exist input data, \"quarter_day\" can passed divide observations classes [0, 6), [6, 12), [12, 18) [18, 24) hour day. default behaviour stratify station (\"SID\") \"quarter_day\".","code":""},{"path":"/reference/check_obs_against_fcst.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Observation error check against forecast — check_obs_against_fcst","text":"object sames class .fcst attribute named removed_cases containing data frame removed cases.","code":""},{"path":"/reference/det_probabilities.html","id":null,"dir":"Reference","previous_headings":"","what":"Compute binary probabilities for deterministic forecasts — det_probabilities","title":"Compute binary probabilities for deterministic forecasts — det_probabilities","text":"Compute binary probabilities deterministic forecasts","code":""},{"path":"/reference/det_probabilities.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compute binary probabilities for deterministic forecasts — det_probabilities","text":"","code":"det_probabilities(.fcst, parameter, thresholds, obs_probabilities = TRUE)"},{"path":"/reference/det_probabilities.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compute binary probabilities for deterministic forecasts — det_probabilities","text":".fcst harp_list object harp_det_point_df deta frames, harp_det_point_df data frame. parameter name column observed data. thresholds numeric vector thresholds compute probabilities. obs_probabilities logical indicating whether compute binary probabilities observations.","code":""},{"path":"/reference/det_probabilities.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compute binary probabilities for deterministic forecasts — det_probabilities","text":"object class .fcst columns threshold, fcst_prob optionally obs_prob instead raw forecast column.","code":""},{"path":"/reference/det_verify.html","id":null,"dir":"Reference","previous_headings":"","what":"Compute verification scores for deterministic forecasts. — det_verify","title":"Compute verification scores for deterministic forecasts. — det_verify","text":"Compute verification scores deterministic forecasts.","code":""},{"path":"/reference/det_verify.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compute verification scores for deterministic forecasts. — det_verify","text":"","code":"det_verify( .fcst, parameter, thresholds = NULL, groupings = \"lead_time\", circle = NULL, hexbin = TRUE, num_bins = 30, show_progress = TRUE, ... ) # S3 method for harp_det_point_df det_verify( .fcst, parameter, thresholds = NULL, groupings = \"lead_time\", circle = NULL, hexbin = TRUE, num_bins = 30, show_progress = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/det_verify.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compute verification scores for deterministic forecasts. — det_verify","text":".fcst harp_df harp_list object tables column observations, single forecast table. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. thresholds numeric vector thresholds compute threshold based scores. Set NULL (default) compute summary scores. groupings groups compute scores. See group_by information grouping works. circle set parameter assumed cyclic bias calculations. distance around circle units parameter, typically value 360 degrees 2 * pi radians. hexbin Logical. Whether compute hexbins forecast, observation pairs. Defaults TRUE. See bin_fcst_obs details. num_bins number bins partition observations hexbin computation. show_progress Logical - whether show progress bars. Defaults TRUE. ... Reserved methods. fcst_model name forecast model use fcst_model column output. function dispatched harp_list object, names harp_list automatically used.","code":""},{"path":"/reference/dplyr_list.html","id":null,"dir":"Reference","previous_headings":"","what":"dplyr verbs for lists — dplyr_list","title":"dplyr verbs for lists — dplyr_list","text":"list data frames, output verification function, may want wrangle data data frames time. can achieved using dplyr verb followed _list. data frames function applicable modified data frame returned. verb fails (e.g. specified columns exist), data frame silently returned unmodified","code":""},{"path":"/reference/dplyr_list.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"dplyr verbs for lists — dplyr_list","text":"","code":"mutate_list(.list, ...) filter_list(.list, ...) select_list(.list, ...)"},{"path":"/reference/dplyr_list.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"dplyr verbs for lists — dplyr_list","text":".list list data frames ... arguments dplyr verb","code":""},{"path":"/reference/dplyr_list.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"dplyr verbs for lists — dplyr_list","text":"list attrbutes input .list","code":""},{"path":[]},{"path":"/reference/ecoval.html","id":null,"dir":"Reference","previous_headings":"","what":"Compute economic value score — ecoval","title":"Compute economic value score — ecoval","text":"Compute economic value score","code":""},{"path":"/reference/ecoval.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compute economic value score — ecoval","text":"","code":"ecoval(obs, pred, costloss, thresholds)"},{"path":"/reference/ecoval.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compute economic value score — ecoval","text":"obs vector observations (value 0,1) pred vector probabilities 0,1. costloss vector cost/loss ratios thresholds vector threshold probabilities","code":""},{"path":"/reference/ecoval.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compute economic value score — ecoval","text":"list cl, value, Vmax, Venv, H, F, s, n","code":""},{"path":"/reference/ens_brier.html","id":null,"dir":"Reference","previous_headings":"","what":"Brier score and its decomposition for an ensemble. — ens_brier","title":"Brier score and its decomposition for an ensemble. — ens_brier","text":"Brier score decomposition ensemble.","code":""},{"path":"/reference/ens_brier.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Brier score and its decomposition for an ensemble. — ens_brier","text":"","code":"ens_brier( .fcst, parameter, thresholds, groupings = \"lead_time\", climatology = \"sample\", rel_probs = NA, num_ref_members = NA, keep_score = c(\"both\", \"brier\", \"reliability\"), show_progress = TRUE, ... ) # S3 method for harp_ens_point_df ens_brier( .fcst, parameter, thresholds, groupings = \"lead_time\", climatology = \"sample\", rel_probs = NA, num_ref_members = NA, keep_score = \"both\", show_progress = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/ens_brier.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Brier score and its decomposition for an ensemble. — ens_brier","text":".fcst harp_df harp_list object tables column observations, single forecast table. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. thresholds numeric vector thresholds compute threshold based scores. Set NULL (default) compute summary scores. groupings groups compute scores. See group_by information grouping works. climatology climatology use Brier Skill Score. Can \"sample\" sample climatology (default), named list elements eps_model member use member eps model harp_fcst object climatology, data frame columns threshold climatology also optionally lead_time. rel_probs Probabilities use reliability diagrams. Set NA (default) select automatically. num_ref_members \"fair\" scores, score scaled valid number ensemble members. Set NA (default) modify score. show_progress Logical - whether show progress bars. Defaults TRUE. ... Reserved methods. fcst_model name forecast model use fcst_model column output. function dispatched harp_list object, names harp_list automatically used.","code":""},{"path":"/reference/ens_brier.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Brier score and its decomposition for an ensemble. — ens_brier","text":"data frame data grouped groupings column(s) columns brier_score, brier_skill_score deomposition brier score - brier_score_reliability, brier_score_resolution brier_score_uncertainty.","code":""},{"path":"/reference/ens_crps.html","id":null,"dir":"Reference","previous_headings":"","what":"Continuous Rank Probability Score (CRPS) for an ensemble. — ens_crps","title":"Continuous Rank Probability Score (CRPS) for an ensemble. — ens_crps","text":"CRPS decomposition computed columns harp_list, harp_ens_grid_df object. Typically scores aggregated lead time, grouping variables cam chosen.","code":""},{"path":"/reference/ens_crps.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Continuous Rank Probability Score (CRPS) for an ensemble. — ens_crps","text":"","code":"ens_crps( .fcst, parameter, groupings = \"lead_time\", num_ref_members = NA, keep_full_output = FALSE, show_progress = TRUE, ... ) # S3 method for harp_ens_point_df ens_crps( .fcst, parameter, groupings = \"lead_time\", num_ref_members = NA, keep_full_output = FALSE, show_progress = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/ens_crps.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Continuous Rank Probability Score (CRPS) for an ensemble. — ens_crps","text":".fcst harp_df harp_list object tables column observations, single forecast table. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. groupings groups compute scores. See group_by information grouping works. num_ref_members \"fair\" scores, score scaled valid number ensemble members. Set NA (default) modify score. show_progress Logical - whether show progress bars. Defaults TRUE. ... Reserved methods. fcst_model name forecast model use fcst_model column output. function dispatched harp_list object, names harp_list automatically used.","code":""},{"path":"/reference/ens_crps.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Continuous Rank Probability Score (CRPS) for an ensemble. — ens_crps","text":"object format inputs data grouped groupings column(s) columns crps, crps_pot crps_rel.","code":""},{"path":"/reference/ens_probabilities.html","id":null,"dir":"Reference","previous_headings":"","what":"Compute probabilities of threshold exceedence for ensemble forecasts — ens_probabilities","title":"Compute probabilities of threshold exceedence for ensemble forecasts — ens_probabilities","text":"Compute probabilities threshold exceedence ensemble forecasts","code":""},{"path":"/reference/ens_probabilities.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compute probabilities of threshold exceedence for ensemble forecasts — ens_probabilities","text":"","code":"ens_probabilities(.fcst, thresholds, parameter = NULL)"},{"path":"/reference/ens_probabilities.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compute probabilities of threshold exceedence for ensemble forecasts — ens_probabilities","text":".fcst harp_df harp_list object tables column observations, single forecast table. thresholds numeric vector thresholds compute threshold based scores. Set NULL (default) compute summary scores. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}.","code":""},{"path":"/reference/ens_probabilities.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compute probabilities of threshold exceedence for ensemble forecasts — ens_probabilities","text":"harp_list object data frame columns threshold, fcst_prob obs_prob instead columns member forecast.","code":""},{"path":"/reference/ens_rank_histogram.html","id":null,"dir":"Reference","previous_headings":"","what":"Rank histogram for an ensemble. — ens_rank_histogram","title":"Rank histogram for an ensemble. — ens_rank_histogram","text":"rank histogram computed columns harp_list object. Typically scores aggregated lead time, grouping variables can chosen.","code":""},{"path":"/reference/ens_rank_histogram.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Rank histogram for an ensemble. — ens_rank_histogram","text":"","code":"ens_rank_histogram( .fcst, parameter, groupings = \"lead_time\", jitter_fcst = NULL, show_progress = TRUE, ... ) # S3 method for harp_ens_point_df ens_rank_histogram( .fcst, parameter, groupings = \"lead_time\", jitter_fcst = NULL, show_progress = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/ens_rank_histogram.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Rank histogram for an ensemble. — ens_rank_histogram","text":".fcst harp_df harp_list object tables column observations, single forecast table. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. groupings groups compute scores. See group_by information grouping works. jitter_fcst function perturb forecast values . used account observation error rank histogram. statistics likely make little difference since expected observations mean error zero. show_progress Logical - whether show progress bars. Defaults TRUE. ... Reserved methods. fcst_model name forecast model use fcst_model column output. function dispatched harp_list object, names harp_list automatically used.","code":""},{"path":"/reference/ens_rank_histogram.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Rank histogram for an ensemble. — ens_rank_histogram","text":"object format inputs data grouped groupings column(s) columns rank rank_count nested together column name rank_histogram.","code":""},{"path":"/reference/ens_read_and_verify.html","id":null,"dir":"Reference","previous_headings":"","what":"Read forecast and observations and verify. — ens_read_and_verify","title":"Read forecast and observations and verify. — ens_read_and_verify","text":"function likely replaced future harp versions since rather cumbersome. thought still needs go finding usable API. wrapper verification process. Forecasts observations read , filtered common cases, errors checked, full verification done scores. minimise memory usage, verification can done one lead time time. also possible parallelise process using example mclapply, future_map.","code":""},{"path":"/reference/ens_read_and_verify.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Read forecast and observations and verify. — ens_read_and_verify","text":"","code":"ens_read_and_verify( start_date, end_date, parameter, fcst_model, fcst_path, obs_path, lead_time = seq(0, 48, 3), num_iterations = length(lead_time), verify_members = TRUE, thresholds = NULL, members = NULL, vertical_coordinate = c(NA_character_, \"pressure\", \"model\", \"height\"), fctable_file_template = \"fctable\", obsfile_template = \"obstable\", groupings = \"lead_time\", by = \"6h\", lags = \"0s\", merge_lags_on_read = TRUE, lag_fcst_models = NULL, parent_cycles = NULL, lag_direction = 1, fcst_shifts = NULL, keep_unshifted = FALSE, drop_neg_leadtimes = TRUE, climatology = \"sample\", stations = NULL, scale_fcst = NULL, scale_obs = NULL, spread_drop_member = NULL, jitter_fcst = NULL, common_cases_only = TRUE, common_cases_xtra_cols = NULL, check_obs_fcst = TRUE, gross_error_check = TRUE, min_allowed = NULL, max_allowed = NULL, num_sd_allowed = NULL, show_progress = FALSE, verif_path = NULL )"},{"path":"/reference/ens_read_and_verify.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Read forecast and observations and verify. — ens_read_and_verify","text":"start_date Start date verification. numeric character. YYYYMMDD(HH)(mm). end_date End date verification. numeric character. parameter parameter verify. fcst_model forecast model(s) verify. Can single string character vector model names. fcst_path path forecast FCTABLE files. obs_path path observation OBSTABLE files. lead_time lead times verify. num_iterations number iterations per verification calculation. default number iterations lead times. small number iterations set, may useful set show_progress = TRUE. higher number iterations, smaller amount data held memory one time. verify_members Whether verify individual members ensemble. Even thresholds supplied, summary scores computed. wish compute categorical scores, separate det_verify function must used. thresholds thresholds compute categorical scores . members members retrieve reading EPS forecast. select members forecast models, numeric vector. specific members specific models named list element name forecast model containing numeric vector. e.g. members = list(eps_model1 = seq(0, 3), eps_model2 = c(2, 3)). multi model ensembles, element named list contain another named list sub model name followed desired members, e.g. members = list(eps_model1 = list(sub_model1 = seq(0, 3), sub_model2 = c(2, 3))) vertical_coordinate vertical co-ordinate. fctable_file_template template file names files read . normally one \"fctable_*\" templates can seen show_file_templates. Can single string, character vector list length fcst_model. named, order templates assumed fcst_model. named, names must match entries fcst_model. obsfile_template template OBSTABLE files - default \"obstable\", OBSTABLE_{YYYY}.sqlite. groupings groups verify . default \"leadtime\". Another common grouping might groupings = c(\"leadtime\", \"fcst_cycle\"). frequency forecast cycles verify. lags lagged forecasts, lags passed read_point_forecast(). merge_lags_on_read Logical. Whether merge lagged ensemble members ensemble. default behaviour. FALSE, lag_forecast used lagging. lag_fcst_models merge_lags_on_read = FALSE, names fcst_models lags applied. parent_cycles merge_lags_on_read = FALSE, parent cycles lagged forecasts. lag_direction merge_lags_on_read = FALSE, direction lagging. 1 Lags backwards time parent cycles -1 lags forwards time. fcst_shifts, keep_unshifted See shift_forecast. drop_neg_leadtimes Logical. Whether drop negative lead times may arise shifting. climatology climatology use Brier Skill Score. Can \"sample\" sample climatology (default), named list elements eps_model member use member eps model harp_fcst object climatology, data frame columns threshold climatology also optionally leadtime. stations stations verify . default use stations station_list common fcst_model domains. scale_fcst named list arguments scale_point_forecast. scale_obs names list arguments scale_point_obs. spread_drop_member members drop calculation ensemble variance standard deviation. harp_fcst objects, can numeric scalar - case recycled forecast models; list numeric vector length harp_fcst object, named list names corresponding names harp_fcst object. jitter_fcst function perturb forecast values . used account observation error rank histogram. statistics likely make little difference since expected observations mean error zero. common_cases_only Logical. Whether select common cases computing verification scores. default TRUE. common_cases_xtra_cols Extra columns use call common_cases check_obs_fcst Logical. Whether check errors observations comparing forecast values. gross_error_check Logical whether perform gross error check. min_allowed minimum value observation allow gross error check. set NULL default value parameter used. max_allowed maximum value observation allow gross error check. set NULL default value parameter used. num_sd_allowed number standard deviations forecast obseravtions within. Set NULL automotic value depeninding parameter. show_progress Logical - whether show progress bar. Defaults FALSE. verif_path set, verification files saved path.","code":""},{"path":"/reference/ens_read_and_verify.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Read forecast and observations and verify. — ens_read_and_verify","text":"list containing two data frames: ens_summary_scores ens_threshold_scores.","code":""},{"path":"/reference/ens_reliability.html","id":null,"dir":"Reference","previous_headings":"","what":"Reliability for an ensemble. — ens_reliability","title":"Reliability for an ensemble. — ens_reliability","text":"Reliability ensemble.","code":""},{"path":"/reference/ens_reliability.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Reliability for an ensemble. — ens_reliability","text":"","code":"ens_reliability( .fcst, parameter, thresholds, groupings = \"lead_time\", climatology = \"sample\", rel_probs = NA, show_progress = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/ens_reliability.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Reliability for an ensemble. — ens_reliability","text":".fcst harp_df harp_list object tables column observations, single forecast table. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. thresholds numeric vector thresholds compute threshold based scores. Set NULL (default) compute summary scores. groupings groups compute scores. See group_by information grouping works. climatology climatology use Brier Skill Score. Can \"sample\" sample climatology (default), named list elements eps_model member use member eps model harp_fcst object climatology, data frame columns threshold climatology also optionally lead_time. rel_probs Probabilities use reliability diagrams. Set NA (default) select automatically. show_progress Logical - whether show progress bars. Defaults TRUE. fcst_model name forecast model use fcst_model column output. function dispatched harp_list object, names harp_list automatically used. ... Reserved methods.","code":""},{"path":"/reference/ens_roc.html","id":null,"dir":"Reference","previous_headings":"","what":"Relative Operating Characteristics for an ensemble. — ens_roc","title":"Relative Operating Characteristics for an ensemble. — ens_roc","text":"Relative Operating Characteristics ensemble.","code":""},{"path":"/reference/ens_roc.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Relative Operating Characteristics for an ensemble. — ens_roc","text":"","code":"ens_roc( .fcst, parameter, thresholds, groupings = \"lead_time\", show_progress = TRUE, ... ) # S3 method for harp_ens_point_df ens_roc( .fcst, parameter, thresholds, groupings = \"lead_time\", show_progress = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/ens_roc.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Relative Operating Characteristics for an ensemble. — ens_roc","text":".fcst harp_df harp_list object tables column observations, single forecast table. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. thresholds numeric vector thresholds compute threshold based scores. Set NULL (default) compute summary scores. groupings groups compute scores. See group_by information grouping works. show_progress Logical - whether show progress bars. Defaults TRUE. ... Reserved methods. fcst_model name forecast model use fcst_model column output. function dispatched harp_list object, names harp_list automatically used.","code":""},{"path":"/reference/ens_roc.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Relative Operating Characteristics for an ensemble. — ens_roc","text":"data frame data grouped groupings column(s) nested column ROC row containing data frame columns: prob forecast probability bin, HR hit rate FAR false alarm rate. Use unnest unnest nested column.","code":""},{"path":"/reference/ens_spread_and_skill.html","id":null,"dir":"Reference","previous_headings":"","what":"Compute the skill (RMSE) and spread of an ensemble forecast — ens_spread_and_skill","title":"Compute the skill (RMSE) and spread of an ensemble forecast — ens_spread_and_skill","text":"ensemble mean spread computed columns harp_list object. Typically scores aggregated lead time grouping variables cam chosen. mean bias also computed.","code":""},{"path":"/reference/ens_spread_and_skill.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compute the skill (RMSE) and spread of an ensemble forecast — ens_spread_and_skill","text":"","code":"ens_spread_and_skill( .fcst, parameter, groupings = \"lead_time\", circle = NULL, spread_drop_member = NULL, jitter_fcst = NULL, show_progress = TRUE, ... ) # S3 method for harp_ens_point_df ens_spread_and_skill( .fcst, parameter, groupings = \"lead_time\", circle = NULL, spread_drop_member = NULL, jitter_fcst = NULL, show_progress = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/ens_spread_and_skill.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compute the skill (RMSE) and spread of an ensemble forecast — ens_spread_and_skill","text":".fcst harp_df harp_list object tables column observations, single forecast table. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. groupings groups compute scores. See group_by information grouping works. circle set parameter assumed cyclic bias calculations. distance around circle units parameter, typically value 360 degrees 2 * pi radians. spread_drop_member members drop calculation ensemble variance standard deviation. harp_fcst objects, can numeric scalar - case recycled forecast models; list numeric vector length harp_fcst object, named list names corresponding names harp_fcst object. jitter_fcst function perturb forecast values . used account observation error rank histogram. statistics likely make little difference since expected observations mean error zero. show_progress Logical - whether show progress bars. Defaults TRUE. ... Reserved methods. fcst_model name forecast model use fcst_model column output. function dispatched harp_list object, names harp_list automatically used.","code":""},{"path":"/reference/ens_spread_and_skill.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compute the skill (RMSE) and spread of an ensemble forecast — ens_spread_and_skill","text":"object format inputs data grouped groupings column(s) columns rmse, spread mean_bias.","code":""},{"path":"/reference/ens_value.html","id":null,"dir":"Reference","previous_headings":"","what":"Economic value for an ensemble. — ens_value","title":"Economic value for an ensemble. — ens_value","text":"Economic value ensemble.","code":""},{"path":"/reference/ens_value.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Economic value for an ensemble. — ens_value","text":"","code":"ens_value( .fcst, parameter, thresholds, groupings = \"lead_time\", show_progress = TRUE, ... ) # S3 method for harp_ens_point_df ens_value( .fcst, parameter, thresholds, groupings = \"lead_time\", show_progress = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/ens_value.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Economic value for an ensemble. — ens_value","text":".fcst harp_df harp_list object tables column observations, single forecast table. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. thresholds numeric vector thresholds compute threshold based scores. Set NULL (default) compute summary scores. groupings groups compute scores. See group_by information grouping works. show_progress Logical - whether show progress bars. Defaults TRUE. ... Reserved methods. fcst_model name forecast model use fcst_model column output. function dispatched harp_list object, names harp_list automatically used.","code":""},{"path":"/reference/ens_value.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Economic value for an ensemble. — ens_value","text":"data frame data grouped groupings column(s) nested column economic value row containing data frame columns: cl cost loss, value economic value. Use unnest unnest nested column.","code":""},{"path":"/reference/ens_verify.html","id":null,"dir":"Reference","previous_headings":"","what":"Compute all verification scores for an ensemble. — ens_verify","title":"Compute all verification scores for an ensemble. — ens_verify","text":"Compute verification scores ensemble.","code":""},{"path":"/reference/ens_verify.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compute all verification scores for an ensemble. — ens_verify","text":"","code":"ens_verify( .fcst, parameter, verify_members = TRUE, thresholds = NULL, groupings = \"lead_time\", circle = NULL, rel_probs = NA, num_ref_members = NA, spread_drop_member = NULL, jitter_fcst = NULL, climatology = \"sample\", hexbin = TRUE, num_bins = 30, rank_hist = TRUE, crps = TRUE, brier = TRUE, roc = TRUE, econ_val = TRUE, show_progress = TRUE, ... ) # S3 method for harp_ens_point_df ens_verify( .fcst, parameter, verify_members = TRUE, thresholds = NULL, groupings = \"lead_time\", circle = NULL, rel_probs = NA, num_ref_members = NA, spread_drop_member = NULL, jitter_fcst = NULL, climatology = \"sample\", hexbin = TRUE, num_bins = 30, rank_hist = TRUE, crps = TRUE, show_progress = TRUE, brier = TRUE, roc = TRUE, econ_val = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/ens_verify.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compute all verification scores for an ensemble. — ens_verify","text":".fcst harp_df harp_list object tables column observations, single forecast table. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. verify_members Whether verify individual members ensemble. Even thresholds supplied, summary scores computed. wish compute categorical scores, separate det_verify function must used. thresholds numeric vector thresholds compute threshold based scores. Set NULL (default) compute summary scores. groupings groups compute scores. See group_by information grouping works. circle set parameter assumed cyclic bias calculations. distance around circle units parameter, typically value 360 degrees 2 * pi radians. rel_probs Probabilities use reliability diagrams. Set NA (default) select automatically. num_ref_members \"fair\" scores, score scaled valid number ensemble members. Set NA (default) modify score. spread_drop_member members drop calculation ensemble variance standard deviation. harp_fcst objects, can numeric scalar - case recycled forecast models; list numeric vector length harp_fcst object, named list names corresponding names harp_fcst object. jitter_fcst function perturb forecast values . used account observation error rank histogram. statistics likely make little difference since expected observations mean error zero. climatology climatology use Brier Skill Score. Can \"sample\" sample climatology (default), named list elements eps_model member use member eps model harp_fcst object climatology, data frame columns threshold climatology also optionally lead_time. hexbin Logical. Whether compute hexbins forecast, observation pairs. Defaults TRUE. See bin_fcst_obs details. num_bins number bins partition observations hexbin computation. rank_hist Logical. Whether compute rank histogram. Defaults TRUE. Note computation rank histogram can slow large number (> 1000) groups. crps Logical. Whether compute CRPS. Defaults TRUE. brier Logical. Whether compute Brier score. Defaults TRUE. ignored thresholds set. roc Logical. Whether compute Relative Operating Characteristic (ROC). Defaults TRUE. ignored thresholds set. econ_val Logical. Whether compute economic value. Defaults TRUE. ignored thresholds set. show_progress Logical - whether show progress bars. Defaults TRUE. ... Reserved methods. fcst_model name forecast model use fcst_model column output. function dispatched harp_list object, names harp_list automatically used.","code":""},{"path":"/reference/ens_verify.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compute all verification scores for an ensemble. — ens_verify","text":"list containting three data frames: ens_summary_scores, ens_threshold_scores det_summary_scores.","code":""},{"path":"/reference/fcprob.html","id":null,"dir":"Reference","previous_headings":"","what":"Compute the forecast probabilities for given thresholds — fcprob","title":"Compute the forecast probabilities for given thresholds — fcprob","text":"Compute forecast probabilities given thresholds","code":""},{"path":"/reference/fcprob.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compute the forecast probabilities for given thresholds — fcprob","text":"","code":"fcprob(fc, thresholds)"},{"path":"/reference/fcprob.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compute the forecast probabilities for given thresholds — fcprob","text":"fc two dimensional array EPS data members columns. byrow thresholds compute probabilities ","code":""},{"path":"/reference/filter.harp_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Filter a harp_fcst object. — filter.harp_fcst","title":"Filter a harp_fcst object. — filter.harp_fcst","text":"Works table harp_fcst object way filter","code":""},{"path":"/reference/filter.harp_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Filter a harp_fcst object. — filter.harp_fcst","text":"","code":"# S3 method for harp_fcst filter(.fcst, ...)"},{"path":"/reference/filter.harp_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Filter a harp_fcst object. — filter.harp_fcst","text":".fcst harp_fcst object. ... Arguments filter","code":""},{"path":"/reference/first_validdate.html","id":null,"dir":"Reference","previous_headings":"","what":"Return the first valid date in a forecast — first_validdate","title":"Return the first valid date in a forecast — first_validdate","text":"version 0.1.0 reading data longer done start date end date, rather vector date-time strings making functions obsolete. unique_valid_dttm now appropriate function use. function intended used find first date fetch observations verify.","code":""},{"path":"/reference/first_validdate.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Return the first valid date in a forecast — first_validdate","text":"","code":"first_validdate(.fcst) last_validdate(.fcst)"},{"path":"/reference/first_validdate.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Return the first valid date in a forecast — first_validdate","text":".fcst harp_fcst object table containing column named validdate data unix time format.","code":""},{"path":"/reference/first_validdate.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Return the first valid date in a forecast — first_validdate","text":"first valid time YYYYMMDDhhmm format input object.","code":""},{"path":"/reference/gather_members.html","id":null,"dir":"Reference","previous_headings":"","what":"Convert EPS forecast data from wide format data frame to long format data\nframe. — gather_members","title":"Convert EPS forecast data from wide format data frame to long format data\nframe. — gather_members","text":"pivot_members now preferred method transforming long wide data frames since supports classes introduced version 0.1.0.","code":""},{"path":"/reference/gather_members.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Convert EPS forecast data from wide format data frame to long format data\nframe. — gather_members","text":"","code":"gather_members( .fcst, member_regex = \"_mbr[[:digit:]]+$|_mbr[[:digit:]]+_lag[[:graph:]]*$\" )"},{"path":"/reference/gather_members.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Convert EPS forecast data from wide format data frame to long format data\nframe. — gather_members","text":".fcst EPS forecast data frame wide format. member_regex Regular expression column names contain forecasts single member.","code":""},{"path":"/reference/gather_members.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Convert EPS forecast data from wide format data frame to long format data\nframe. — gather_members","text":"EPS data frame long format.","code":""},{"path":"/reference/harpPoint-package.html","id":null,"dir":"Reference","previous_headings":"","what":"harpPoint: Point verifition for NWP forecasts — harpPoint-package","title":"harpPoint: Point verifition for NWP forecasts — harpPoint-package","text":"Functions computing verification scores NWP forecasts. Part harp ecosystem.","code":""},{"path":[]},{"path":"/reference/harpPoint-package.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"harpPoint: Point verifition for NWP forecasts — harpPoint-package","text":"Maintainer: Andrew Singleton andrewts@met.Authors: Alex Deckmyn alex.deckmyn@meteo.","code":""},{"path":"/reference/jitter_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Jitter a forecast to account for observation errors. — jitter_fcst","title":"Jitter a forecast to account for observation errors. — jitter_fcst","text":"account observation errors ensemble verification forecast values can perturbed sampling specified error distribution. Jittering forecast likely effect ensemble spread rank histograms.","code":""},{"path":"/reference/jitter_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Jitter a forecast to account for observation errors. — jitter_fcst","text":"","code":"jitter_fcst(.fcst, jitter_func, obs_col = NULL, ...)"},{"path":"/reference/jitter_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Jitter a forecast to account for observation errors. — jitter_fcst","text":".fcst object class 'harp_fcst'. jitter_func function applied forecast values. obs_col observations column used jitter function. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. ... arguments jitter_func.","code":""},{"path":"/reference/jitter_fcst.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Jitter a forecast to account for observation errors. — jitter_fcst","text":"object class .fcst jittered forecast values.","code":""},{"path":"/reference/jitter_fcst.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Jitter a forecast to account for observation errors. — jitter_fcst","text":"Note jitter function function works vector forecast values one two arguments. first argument forecast data, second argument, observations.","code":""},{"path":"/reference/join_models.html","id":null,"dir":"Reference","previous_headings":"","what":"Join all models into a single ensemble. — join_models","title":"Join all models into a single ensemble. — join_models","text":"function useful finding common cases models.","code":""},{"path":"/reference/join_models.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Join all models into a single ensemble. — join_models","text":"","code":"join_models(.fcst, join_type = \"inner\", name = \"joined_models\", ...)"},{"path":"/reference/join_models.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Join all models into a single ensemble. — join_models","text":".fcst harp_list object multimodel data merged merge_multimodel. join_type type join perform. See join. name name resulting model. ... arguments join.","code":""},{"path":"/reference/join_models.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Join all models into a single ensemble. — join_models","text":"harp_df data frame.","code":""},{"path":"/reference/lag_forecast.html","id":null,"dir":"Reference","previous_headings":"","what":"Lags a forecast — lag_forecast","title":"Lags a forecast — lag_forecast","text":"Lagging done supplying parent forecast cycles. function work cycles belong parent return lagged forecast members child cycles appended parent cycle.","code":""},{"path":"/reference/lag_forecast.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Lags a forecast — lag_forecast","text":"","code":"lag_forecast(.fcst, fcst_model, parent_cycles, direction = 1)"},{"path":"/reference/lag_forecast.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Lags a forecast — lag_forecast","text":".fcst harp_fcst object. fcst_model name forecast model harp_fcst object lagged. Must quoted. parent_cycles numeric vector forecast cycles form parents lagging. Members parent cycles child cycles. dierction direction lagging . 1 (default) lags backwards time parent cycles -1 lags forwards time.","code":""},{"path":"/reference/lag_forecast.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Lags a forecast — lag_forecast","text":"harp_fcst object fcst_model now containing lagged forecast.","code":""},{"path":"/reference/make_bootstrap_pools.html","id":null,"dir":"Reference","previous_headings":"","what":"Make pools for block bootstrapping — make_bootstrap_pools","title":"Make pools for block bootstrapping — make_bootstrap_pools","text":"block bootstrap using bootstrap_verify, blocks can passed data frame \"pool\" column telling bootstrap_verify pool data blocks. make_bootstrap_pools function make data frame.","code":""},{"path":"/reference/make_bootstrap_pools.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Make pools for block bootstrapping — make_bootstrap_pools","text":"","code":"make_bootstrap_pools(.fcst, pool_col, pool_length, overlap = FALSE)"},{"path":"/reference/make_bootstrap_pools.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Make pools for block bootstrapping — make_bootstrap_pools","text":".fcst harp_fcst object pool_col column used define pools. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}} pool_length length pool. Numeric character unit qualifier pool_col date-time format. unit qualifier can : \"s\" = seconds, \"m\" = minutes, \"h\" = hours, \"d\" = days. overlap Logical. Whether pools overlap.","code":""},{"path":"/reference/make_bootstrap_pools.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Make pools for block bootstrapping — make_bootstrap_pools","text":"data frame columns pool_col \"pool\".","code":""},{"path":"/reference/make_bootstrap_pools.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Make pools for block bootstrapping — make_bootstrap_pools","text":"Typically block bootstrapping used serial auto-correlations data. example auto-correlations suspected forecasts, pools defined fcdate column create blocks data auto-correlations maintained. Pools may set overlap, whereby new pool created beginning new value pool_col. length pool defined units used pool_col - pool_col date-time column, pool_length assumed hours, though units can set adding qualifier letter: \"s\" = seconds, \"m\" = minutes, \"h\" = hours, \"d\" = days.","code":""},{"path":"/reference/make_bootstrap_pools.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Make pools for block bootstrapping — make_bootstrap_pools","text":"","code":"make_bootstrap_pools(ens_point_df, lead_time, 2) #> # A tibble: 24 × 2 #> lead_time pool #>
#> 1 0 1 #> 2 1 1 #> 3 2 2 #> 4 3 2 #> 5 4 3 #> 6 5 3 #> 7 6 4 #> 8 7 4 #> 9 8 5 #> 10 9 5 #> # ℹ 14 more rows make_bootstrap_pools(ens_point_df, lead_time, 2, overlap = TRUE) #> # A tibble: 46 × 2 #> lead_time pool #> #> 1 0 1 #> 2 1 1 #> 3 1 2 #> 4 2 2 #> 5 2 3 #> 6 3 3 #> 7 3 4 #> 8 4 4 #> 9 4 5 #> 10 5 5 #> # ℹ 36 more rows # pool_col as a variable my_col <- \"lead_time\" make_bootstrap_pools(ens_point_df, {{my_col}}, 2) #> # A tibble: 24 × 2 #> lead_time pool #> #> 1 0 1 #> 2 1 1 #> 3 2 2 #> 4 3 2 #> 5 4 3 #> 6 5 3 #> 7 6 4 #> 8 7 4 #> 9 8 5 #> 10 9 5 #> # ℹ 14 more rows"},{"path":"/reference/merge_multimodel.html","id":null,"dir":"Reference","previous_headings":"","what":"Merge sub models of a multimodel ensemble into a single ensemble — merge_multimodel","title":"Merge sub models of a multimodel ensemble into a single ensemble — merge_multimodel","text":"multimodel ensembles read sub models stored. whole ensemble needed, function merge sub models single ensemble. Note renaming members sub model names retained. multi model ensembles, input silently returned unaltered.","code":""},{"path":"/reference/merge_multimodel.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Merge sub models of a multimodel ensemble into a single ensemble — merge_multimodel","text":"","code":"merge_multimodel(.fcst, keep_sub_models = TRUE)"},{"path":"/reference/merge_multimodel.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Merge sub models of a multimodel ensemble into a single ensemble — merge_multimodel","text":".fcst object class harp_fcst read read_point_forecast keep_sub_models Set FALSE discard sub models separate elements harp_fcst list. default behaviour keep .","code":""},{"path":"/reference/merge_multimodel.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Merge sub models of a multimodel ensemble into a single ensemble — merge_multimodel","text":"harp_fcst object one layer - element table forecast data model.","code":""},{"path":"/reference/mutate.harp_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Mutate columns from tables in a harp_fcst object. — mutate.harp_fcst","title":"Mutate columns from tables in a harp_fcst object. — mutate.harp_fcst","text":"Works way mutate, except runs tables harp_fcst object.","code":""},{"path":"/reference/mutate.harp_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Mutate columns from tables in a harp_fcst object. — mutate.harp_fcst","text":"","code":"# S3 method for harp_fcst mutate(.fcst, ...)"},{"path":"/reference/mutate.harp_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Mutate columns from tables in a harp_fcst object. — mutate.harp_fcst","text":".fcst harp_fcst object ... Arguments mutate","code":""},{"path":"/reference/mutate_at.harp_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Mutate selected columns from tables in a harp_fcst object. — mutate_at.harp_fcst","title":"Mutate selected columns from tables in a harp_fcst object. — mutate_at.harp_fcst","text":"Works way mutate_at, except runs tables harp_fcst object.","code":""},{"path":"/reference/mutate_at.harp_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Mutate selected columns from tables in a harp_fcst object. — mutate_at.harp_fcst","text":"","code":"mutate_at.harp_fcst(.fcst, .mutate_vars, .mutate_funs, ...)"},{"path":"/reference/mutate_at.harp_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Mutate selected columns from tables in a harp_fcst object. — mutate_at.harp_fcst","text":".fcst harp_fcst object ... Arguments mutate","code":""},{"path":"/reference/perfect_score.html","id":null,"dir":"Reference","previous_headings":"","what":"Set or modify and output the perfect scores for different verificaiton\nmetrics — perfect_score","title":"Set or modify and output the perfect scores for different verificaiton\nmetrics — perfect_score","text":"function returns named vector giving perfect score verification scores computed harpPoint give single value. perfect values can modified specifying named arguments perfect scores new scores can set way.","code":""},{"path":"/reference/perfect_score.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Set or modify and output the perfect scores for different verificaiton\nmetrics — perfect_score","text":"","code":"perfect_score(...)"},{"path":"/reference/perfect_score.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Set or modify and output the perfect scores for different verificaiton\nmetrics — perfect_score","text":"... Named arguments give values perfect score","code":""},{"path":"/reference/perfect_score.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Set or modify and output the perfect scores for different verificaiton\nmetrics — perfect_score","text":"named vector","code":""},{"path":"/reference/perfect_score.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Set or modify and output the perfect scores for different verificaiton\nmetrics — perfect_score","text":"","code":"perfect_score() #> bias rmse mae #> 0e+00 0e+00 0e+00 #> stde threat_score hit_rate #> 0e+00 1e+00 1e+00 #> miss_rate false_alarm_rate false_alarm_ratio #> 0e+00 0e+00 0e+00 #> heidke_skill_score pierce_skill_score kuiper_skill_score #> 1e+00 1e+00 1e+00 #> percent_correct frequency_bias equitable_threat_score #> 1e+00 1e+00 1e+00 #> odds_ratio log_odds_ratio odds_ratio_skill_score #> 1e+06 1e+06 1e+00 #> extreme_dependency_score symmetric_eds extreme_dependency_index #> 1e+00 1e+00 1e+00 #> symmetric_edi mean_bias spread #> 1e+00 0e+00 1e+06 #> spread_skill_ratio crps crps_potential #> 1e+00 0e+00 0e+00 #> crps_reliability fair_brier_score fair_crps #> 0e+00 0e+00 0e+00 #> brier_score brier_skill_score brier_score_reliability #> 0e+00 1e+00 0e+00 #> brier_score_resolution roc_area #> 1e+00 1e+00 perfect_score(bss = 1) #> bias rmse mae #> 0e+00 0e+00 0e+00 #> stde threat_score hit_rate #> 0e+00 1e+00 1e+00 #> miss_rate false_alarm_rate false_alarm_ratio #> 0e+00 0e+00 0e+00 #> heidke_skill_score pierce_skill_score kuiper_skill_score #> 1e+00 1e+00 1e+00 #> percent_correct frequency_bias equitable_threat_score #> 1e+00 1e+00 1e+00 #> odds_ratio log_odds_ratio odds_ratio_skill_score #> 1e+06 1e+06 1e+00 #> extreme_dependency_score symmetric_eds extreme_dependency_index #> 1e+00 1e+00 1e+00 #> symmetric_edi mean_bias spread #> 1e+00 0e+00 1e+06 #> spread_skill_ratio crps crps_potential #> 1e+00 0e+00 0e+00 #> crps_reliability fair_brier_score fair_crps #> 0e+00 0e+00 0e+00 #> brier_score brier_skill_score brier_score_reliability #> 0e+00 1e+00 0e+00 #> brier_score_resolution roc_area bss #> 1e+00 1e+00 1e+00 perfect_score(bias = -1) #> bias rmse mae #> -1e+00 0e+00 0e+00 #> stde threat_score hit_rate #> 0e+00 1e+00 1e+00 #> miss_rate false_alarm_rate false_alarm_ratio #> 0e+00 0e+00 0e+00 #> heidke_skill_score pierce_skill_score kuiper_skill_score #> 1e+00 1e+00 1e+00 #> percent_correct frequency_bias equitable_threat_score #> 1e+00 1e+00 1e+00 #> odds_ratio log_odds_ratio odds_ratio_skill_score #> 1e+06 1e+06 1e+00 #> extreme_dependency_score symmetric_eds extreme_dependency_index #> 1e+00 1e+00 1e+00 #> symmetric_edi mean_bias spread #> 1e+00 0e+00 1e+06 #> spread_skill_ratio crps crps_potential #> 1e+00 0e+00 0e+00 #> crps_reliability fair_brier_score fair_crps #> 0e+00 0e+00 0e+00 #> brier_score brier_skill_score brier_score_reliability #> 0e+00 1e+00 0e+00 #> brier_score_resolution roc_area #> 1e+00 1e+00"},{"path":"/reference/pipe.html","id":null,"dir":"Reference","previous_headings":"","what":"Pipe operator — %>%","title":"Pipe operator — %>%","text":"See magrittr::%>% details.","code":""},{"path":"/reference/pipe.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Pipe operator — %>%","text":"","code":"lhs %>% rhs"},{"path":[]},{"path":"/reference/pooled_bootstrap_score.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Bootstrapping by pools — pooled_bootstrap_score","text":"","code":"pooled_bootstrap_score( .fcst, score_function, parameter, n, pooled_by = \"fcdate\", groupings = \"leadtime\", confidence_interval = 0.95, min_cases = 25, perfect_scores = perfect_score(), show_progress = TRUE, ... )"},{"path":"/reference/pull.harp_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Pull columns from tables in a harp_fcst object. — pull.harp_fcst","title":"Pull columns from tables in a harp_fcst object. — pull.harp_fcst","text":"Works way pull, except runs tables harp_fcst object. means common columns objects can safely pulled.","code":""},{"path":"/reference/pull.harp_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Pull columns from tables in a harp_fcst object. — pull.harp_fcst","text":"","code":"# S3 method for harp_fcst pull(.fcst, ...)"},{"path":"/reference/pull.harp_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Pull columns from tables in a harp_fcst object. — pull.harp_fcst","text":".fcst harp_fcst object ... Arguments pull","code":""},{"path":"/reference/pull_stations.html","id":null,"dir":"Reference","previous_headings":"","what":"Pull station IDs from forecast or observations — pull_stations","title":"Pull station IDs from forecast or observations — pull_stations","text":"Pull station IDs forecast observations","code":""},{"path":"/reference/pull_stations.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Pull station IDs from forecast or observations — pull_stations","text":"","code":"pull_stations(.fcst)"},{"path":"/reference/pull_stations.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Pull station IDs from forecast or observations — pull_stations","text":".fcst harp_fcst object data frame","code":""},{"path":"/reference/pull_stations.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Pull station IDs from forecast or observations — pull_stations","text":"vector station IDs","code":""},{"path":"/reference/qair2rh.html","id":null,"dir":"Reference","previous_headings":"","what":"qair2rh — qair2rh","title":"qair2rh — qair2rh","text":"Convert specific humidity relative humidity","code":""},{"path":"/reference/qair2rh.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"qair2rh — qair2rh","text":"","code":"qair2rh(qair, temp, press = 1013.25)"},{"path":"/reference/qair2rh.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"qair2rh — qair2rh","text":"qair specific humidity, dimensionless (e.g. kg/kg) ratio water mass / total air mass temp degrees C press pressure mb","code":""},{"path":"/reference/qair2rh.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"qair2rh — qair2rh","text":"rh relative humidity, ratio actual water mixing ratio saturation mixing ratio","code":""},{"path":"/reference/qair2rh.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"qair2rh — qair2rh","text":"converting specific humidity relative humidity NCEP surface flux data RH Bolton 1980 computation Equivalent Potential Temperature http://www.eol.ucar.edu/projects/ceop/dm/documents/refdata_report/eqns.html function lifetd data.atmosphere package https://github.com/PecanProject/pecan/blob/master/modules/data.atmosphere/R/metutils.R","code":""},{"path":"/reference/qair2rh.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"qair2rh — qair2rh","text":"David LeBauer","code":""},{"path":"/reference/rankHistogram.html","id":null,"dir":"Reference","previous_headings":"","what":"Compute the rank histogram for an EPS — rankHistogram","title":"Compute the rank histogram for an EPS — rankHistogram","text":"Compute rank histogram EPS","code":""},{"path":"/reference/rankHistogram.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compute the rank histogram for an EPS — rankHistogram","text":"","code":"rankHistogram(obs, fc)"},{"path":"/reference/rankHistogram.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compute the rank histogram for an EPS — rankHistogram","text":"obs vector observations. fc two dimensional array EPS data members columns.","code":""},{"path":"/reference/reexports.html","id":null,"dir":"Reference","previous_headings":"","what":"Objects exported from other packages — reexports","title":"Objects exported from other packages — reexports","text":"objects imported packages. Follow links see documentation. dplyr vars","code":""},{"path":"/reference/rename.harp_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Rename columns from tables in a harp_fcst object. — rename.harp_fcst","title":"Rename columns from tables in a harp_fcst object. — rename.harp_fcst","text":"Works way rename, except runs tables harp_fcst object. means common columns objects can safely renamed.","code":""},{"path":"/reference/rename.harp_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Rename columns from tables in a harp_fcst object. — rename.harp_fcst","text":"","code":"# S3 method for harp_fcst rename(.fcst, ...)"},{"path":"/reference/rename.harp_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Rename columns from tables in a harp_fcst object. — rename.harp_fcst","text":".fcst harp_fcst object ... Arguments rename","code":""},{"path":"/reference/rename_with.harp_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Rename multiple columns from tables in a harp_fcst object. — rename_with.harp_fcst","title":"Rename multiple columns from tables in a harp_fcst object. — rename_with.harp_fcst","text":"Works way rename_with, except runs tables harp_fcst object. means common columns objects can safely renamed.","code":""},{"path":"/reference/rename_with.harp_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Rename multiple columns from tables in a harp_fcst object. — rename_with.harp_fcst","text":"","code":"# S3 method for harp_fcst rename_with(.fcst, ...)"},{"path":"/reference/rename_with.harp_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Rename multiple columns from tables in a harp_fcst object. — rename_with.harp_fcst","text":".fcst harp_fcst object ... Arguments rename_with","code":""},{"path":"/reference/roc.html","id":null,"dir":"Reference","previous_headings":"","what":"Compute ROC and area under ROC — roc","title":"Compute ROC and area under ROC — roc","text":"Compute ROC area ROC","code":""},{"path":"/reference/roc.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compute ROC and area under ROC — roc","text":"","code":"roc(obs, pred, thresholds)"},{"path":"/reference/roc.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compute ROC and area under ROC — roc","text":"obs vector observations (value 0,1) pred vector probabilities 0,1. thresholds vector threshold probabilities","code":""},{"path":"/reference/roc.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compute ROC and area under ROC — roc","text":"list area vectors thresholds, H, F","code":""},{"path":"/reference/scale_point_forecast.html","id":null,"dir":"Reference","previous_headings":"","what":"Scale forecast data — scale_point_forecast","title":"Scale forecast data — scale_point_forecast","text":"scale_param now preferred function since works point gridded data. wish scale forecast values, example temperature data Kelvin want degrees C, function can used scale data.","code":""},{"path":"/reference/scale_point_forecast.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Scale forecast data — scale_point_forecast","text":"","code":"scale_point_forecast( .fcst, scale_factor, new_units = NULL, multiplicative = FALSE )"},{"path":"/reference/scale_point_forecast.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Scale forecast data — scale_point_forecast","text":".fcst harp_fcst object read read_point_forecast harpIO. scale_factor scaling factor. new_units name new units - set NULL, name changed. multiplicative scaling done multiplicatively, .e. new forecast value old value * scale_factor, set multiplicative TRUE. default (multiplicative = FALSE) scaling additively, .e. new forecast value old value + scale_factor.","code":""},{"path":"/reference/scale_point_forecast.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Scale forecast data — scale_point_forecast","text":"harp_fcst object forecast values scaled scale_factor.","code":""},{"path":"/reference/scale_point_obs.html","id":null,"dir":"Reference","previous_headings":"","what":"Scale observations data — scale_point_obs","title":"Scale observations data — scale_point_obs","text":"scale_param now preferred function since works point gridded data. wish scale observations values, example temperature data Kelvin want degrees C, function can used scale data.","code":""},{"path":"/reference/scale_point_obs.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Scale observations data — scale_point_obs","text":"","code":"scale_point_obs( .obs, parameter, scale_factor, new_units = NULL, multiplicative = FALSE )"},{"path":"/reference/scale_point_obs.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Scale observations data — scale_point_obs","text":".obs data frame point observations read read read_point_obs harpIO. parameter column name data scaled. Must unquoted. scale_factor scaling factor. new_units name new units - set NULL, name changed. multiplicative scaling done multiplicatively, .e. new forecast value old value * scale_factor, set multiplicative TRUE. default (multiplicative = FALSE) scaling additively, .e. new forecast value old value + scale_factor.","code":""},{"path":"/reference/scale_point_obs.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Scale observations data — scale_point_obs","text":"observations data frame parameter column scaled scale_factor.","code":""},{"path":"/reference/select.harp_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Select columns from tables in a harp_fcst object. — select.harp_fcst","title":"Select columns from tables in a harp_fcst object. — select.harp_fcst","text":"Works way select, except runs tables harp_fcst object. means common columns objects can safely selected.","code":""},{"path":"/reference/select.harp_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Select columns from tables in a harp_fcst object. — select.harp_fcst","text":"","code":"# S3 method for harp_fcst select(.fcst, ...)"},{"path":"/reference/select.harp_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Select columns from tables in a harp_fcst object. — select.harp_fcst","text":".fcst harp_fcst object ... Arguments select","code":""},{"path":"/reference/shift_forecast.html","id":null,"dir":"Reference","previous_headings":"","what":"Shift a forecast — shift_forecast","title":"Shift a forecast — shift_forecast","text":"function used shift start times lead times forecasts simulate lagging.","code":""},{"path":"/reference/shift_forecast.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Shift a forecast — shift_forecast","text":"","code":"shift_forecast( .fcst, fcst_shifts, keep_unshifted = FALSE, drop_negative_lead_times = TRUE )"},{"path":"/reference/shift_forecast.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Shift a forecast — shift_forecast","text":".fcst harp_fcst object created read_point_forecast, data frame columns including 'fcdate' (seconds) 'leadtime' (hours). fcst_shifts named list names exist '.fcst' single numeric value apply forecast models. list, element must numeric length 1 1 shift can applied forecast. shifts specified hours. Postive values shift forecast start dates forward time reduce lead times corresponding amounts. Negative values opposite.","code":""},{"path":"/reference/shift_forecast.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Shift a forecast — shift_forecast","text":"object class '.fcst' forecast start times lead times shifted forecast models number hours given 'fcst_shifts'.","code":""},{"path":"/reference/sort_members.html","id":null,"dir":"Reference","previous_headings":"","what":"Sort a 2d array. — sort_members","title":"Sort a 2d array. — sort_members","text":"Sort 2d array.","code":""},{"path":"/reference/sort_members.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Sort a 2d array. — sort_members","text":"","code":"sort_members(x, byrow = TRUE)"},{"path":"/reference/sort_members.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Sort a 2d array. — sort_members","text":"x two dimensional numeric array. byrow Set true sort rows, false sort columns.","code":""},{"path":"/reference/spread_members.html","id":null,"dir":"Reference","previous_headings":"","what":"Convert EPS forecast data from long format data frame to wide format data\nframe. — spread_members","title":"Convert EPS forecast data from long format data frame to wide format data\nframe. — spread_members","text":"pivot_members now preferred method transforming long wide data frames since supports classes introduced version 0.1.0.","code":""},{"path":"/reference/spread_members.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Convert EPS forecast data from long format data frame to wide format data\nframe. — spread_members","text":"","code":"spread_members(.fcst, ...)"},{"path":"/reference/spread_members.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Convert EPS forecast data from long format data frame to wide format data\nframe. — spread_members","text":".fcst EPS forecast data frame long format.","code":""},{"path":"/reference/spread_members.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Convert EPS forecast data from long format data frame to wide format data\nframe. — spread_members","text":"EPS data frame wide format.","code":""},{"path":"/reference/transmute.harp_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Transmute columns from tables in a harp_fcst object. — transmute.harp_fcst","title":"Transmute columns from tables in a harp_fcst object. — transmute.harp_fcst","text":"Works way transmute, except runs tables harp_fcst object. means common columns objects can safely arranged.","code":""},{"path":"/reference/transmute.harp_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Transmute columns from tables in a harp_fcst object. — transmute.harp_fcst","text":"","code":"# S3 method for harp_fcst transmute(.fcst, ...)"},{"path":"/reference/transmute.harp_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Transmute columns from tables in a harp_fcst object. — transmute.harp_fcst","text":".fcst harp_fcst object ... Arguments transmute","code":""}]
+[{"path":"/LICENSE.html","id":null,"dir":"","previous_headings":"","what":"MIT License","title":"MIT License","text":"Copyright (c) 2018 Andrew Singleton Permission hereby granted, free charge, person obtaining copy software associated documentation files (“Software”), deal Software without restriction, including without limitation rights use, copy, modify, merge, publish, distribute, sublicense, /sell copies Software, permit persons Software furnished , subject following conditions: copyright notice permission notice shall included copies substantial portions Software. SOFTWARE PROVIDED “”, WITHOUT WARRANTY KIND, EXPRESS IMPLIED, INCLUDING LIMITED WARRANTIES MERCHANTABILITY, FITNESS PARTICULAR PURPOSE NONINFRINGEMENT. EVENT SHALL AUTHORS COPYRIGHT HOLDERS LIABLE CLAIM, DAMAGES LIABILITY, WHETHER ACTION CONTRACT, TORT OTHERWISE, ARISING , CONNECTION SOFTWARE USE DEALINGS SOFTWARE.","code":""},{"path":"/authors.html","id":null,"dir":"","previous_headings":"","what":"Authors","title":"Authors and Citation","text":"Andrew Singleton. Author, maintainer. Alex Deckmyn. Author.","code":""},{"path":"/authors.html","id":"citation","dir":"","previous_headings":"","what":"Citation","title":"Authors and Citation","text":"Singleton , Deckmyn (2023). harpPoint: Point verifition NWP forecasts. R package version 0.2.0, https://github.com/harphub/harpPoint.","code":"@Manual{, title = {harpPoint: Point verifition for NWP forecasts}, author = {Andrew Singleton and Alex Deckmyn}, year = {2023}, note = {R package version 0.2.0}, url = {https://github.com/harphub/harpPoint}, }"},{"path":"/index.html","id":"harppoint-","dir":"","previous_headings":"","what":"Point verifition for NWP forecasts","title":"Point verifition for NWP forecasts","text":"harpPoint provides functionality verification meteorological data geographic points. Typically verification forecasts interpolated locations weather stations. Functions provided computing verification scores deterministic ensemble forecasts. addition, confidence intervals scores, differences scores different forecast models can computed using bootstrapping.","code":""},{"path":"/index.html","id":"installation","dir":"","previous_headings":"","what":"Installation","title":"Point verifition for NWP forecasts","text":"can install harpPoint GitHub :","code":"# install.packages(\"remotes\") remotes::install_github(\"harphub/harpPoint\")"},{"path":"/index.html","id":"verification","dir":"","previous_headings":"","what":"Verification","title":"Point verifition for NWP forecasts","text":"harpPoint functions verification designed work data read using functions harpIO. means harp_df data frames harp_lists. data must include column observations forecasts verified. two main functions verification: det_verify() - deterministic forecasts; ens_verify() - ensemble forecasts. functions output harp_verif list. list data frames scores separated summary scores threshold scores. Threshold scores computed thresholds provided functions computed probabilities threshold exceedance.","code":""},{"path":"/index.html","id":"deterministic-scores","dir":"","previous_headings":"","what":"Deterministic scores","title":"Point verifition for NWP forecasts","text":"det_verify() computes scores following column names:","code":""},{"path":"/index.html","id":"summary-scores","dir":"","previous_headings":"Deterministic scores","what":"Summary scores","title":"Point verifition for NWP forecasts","text":"bias - mean difference forecasts observations rmse - root mean squared error mae - mean absolute error stde - standard deviation error hexbin - heat map paired hexagonal bins forecasts observations","code":""},{"path":"/index.html","id":"threshold-scores","dir":"","previous_headings":"Deterministic scores","what":"Threshold scores","title":"Point verifition for NWP forecasts","text":"cont_tab - contingency table forecast hits, misses, false alarms correct rejections threat_score - ratio hits sum hits, misses false alarms hit_rate - ratio hits sum hits misses miss_rate - ratio misses sum hits misses false_alarm_rate - ratio false alarms sum false alarms correct rejections false_alarm_ratio - ratio false alarms sum false alarms hits heidke_skill_score - fraction correct forecasts eliminating forecasts correct purely due random chance pierce_skill_score - 1 - miss rate - false alarm rate kuiper_skill_score - well forecasts separates hits false alarms percent_correct - ratio sum hits correct rejections total number cases frequency_bias - ratio sum hits false alarms sum hits misses equitable_threat_score - well forecast measures hits accounting hits due pure chance odds_ratio - ratio product hits correct rejections product misses false alarms log_odds_ratio - sum logs hits correct rejections minus sum logs misses false alarms. odds_ratio_skill_score - ratio product hits correct rejections minus product misses false alarms product hits correct rejections plus product misses false alarms extreme_dependency_score - ratio difference logs observations climatology hit rate sum logs observations climatology hit rate symmetric_eds - symmetric extreme dependency score, ratio difference logs forecast climatology hit rate sum logs forecast climatology hit rate extreme_dependency_index - ratio difference logs false alarm rate hit rate sum logs false alarm rate hit rate symmetric_edi - symmetric extreme dependency index, ratio sum difference logs false alarm rate hit rate difference logs inverse hit rate false alarm rate sum logs hit rate, false alarm rate, inverse hit rate inverse false alarm rate. inverse 1 - value.","code":""},{"path":"/index.html","id":"ensemble-scores","dir":"","previous_headings":"","what":"Ensemble scores","title":"Point verifition for NWP forecasts","text":"ens_verify() computes scores following column names:","code":""},{"path":"/index.html","id":"summary-scores-1","dir":"","previous_headings":"Ensemble scores","what":"Summary scores","title":"Point verifition for NWP forecasts","text":"mean_bias - mean difference ensemble mean forecasts observations rmse - root mean squared error stde - standard deviation error spread - square root mean variance ensemble forecasts hexbin - heat map paired hexagonal bins forecasts observations rank_histogram - Observation counts ranked ensemble member bins crps- cumulative rank probability score - difference cumulative distribution ensemble forecasts step function observations crps_potential - crps achieved perfectly reliable ensemble crps_reliability - Measures ability ensemble produce cumulative distribution desired statisical properties. fair_crps - crps achieved either ensemble infinite number members, number members provided function","code":""},{"path":"/index.html","id":"threshold-scores-1","dir":"","previous_headings":"Ensemble scores","what":"Threshold scores","title":"Point verifition for NWP forecasts","text":"brier_score - mean squared error ensemble probability space fair_brier_score - Brier score achieved either ensemble infinite number members, number members provided function brier_skill_score - Brier score compared reference probabilistic forecast (usually observed climatology) brier_score_reliability - measure ensemble’s ability produce reliable (forecast probability = observed frequency) forecasts brier_score_resolution - measure ensemble’s ability discriminate “day” uncertainty climatological uncertainty brier_score_uncertainty - inherent uncertainty events reliability - frequency observations bins forecast probability roc - relative operating characteristic ensemble - hit rates false alarm rates forecast probability bins roc_area - area roc curve - summarises ability ensemble discriminate events non events economic_value - relative improvement economic value forecast compared climatology range cost / loss ratios","code":""},{"path":"/index.html","id":"getting-gridded-data-to-points","dir":"","previous_headings":"","what":"Getting gridded data to points","title":"Point verifition for NWP forecasts","text":"interpolation gridded data points see Interpolate section Transforming model data article harpIO website, documentation geo_points.","code":""},{"path":"/reference/arrange.harp_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Arrange columns from tables in a harp_fcst object. — arrange.harp_fcst","title":"Arrange columns from tables in a harp_fcst object. — arrange.harp_fcst","text":"Works way arrange, except runs tables harp_fcst object. means common columns objects can safely arranged.","code":""},{"path":"/reference/arrange.harp_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Arrange columns from tables in a harp_fcst object. — arrange.harp_fcst","text":"","code":"# S3 method for harp_fcst arrange(.fcst, ...)"},{"path":"/reference/arrange.harp_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Arrange columns from tables in a harp_fcst object. — arrange.harp_fcst","text":".fcst harp_fcst object ... Arguments arrange","code":""},{"path":"/reference/bin_fcst_obs.html","id":null,"dir":"Reference","previous_headings":"","what":"Create a bi-variate histogram of forecast, observation pairs — bin_fcst_obs","title":"Create a bi-variate histogram of forecast, observation pairs — bin_fcst_obs","text":"Values forecasts observations binned bands whereby density forecast, observation pairs bin calculated. hood, data binned hexagons using hexbin. Hexagons used since symmetry nearest neighbours unlike square bins, plot time polygon maximum number sides tessellate.","code":""},{"path":"/reference/bin_fcst_obs.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Create a bi-variate histogram of forecast, observation pairs — bin_fcst_obs","text":"","code":"bin_fcst_obs( .fcst, parameter, groupings = \"lead_time\", num_bins = 30, show_progress = TRUE, ... ) # S3 method for harp_det_point_df bin_fcst_obs( .fcst, parameter, groupings = \"lead_time\", num_bins = 30, show_progress = TRUE, fcst_model = NULL, ... ) # S3 method for harp_ens_point_df bin_fcst_obs( .fcst, parameter, groupings = \"lead_time\", num_bins = 30, show_progress = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/bin_fcst_obs.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Create a bi-variate histogram of forecast, observation pairs — bin_fcst_obs","text":".fcst harp_df data frame harp_list. parameter column containing parameter. Can unquoted, quoted string, embraced variable name (.e var). groupings groupings compute binned densities. Must vector strings, list vectors strings. num_bins number bins partition observations. show_progress Logical. Whether show progress bar. ... Arguments methods. fcst_model name forecast model. .fcst contain fcst_model, new column created populated value. fcst_model column exists, value column replaced value.","code":""},{"path":"/reference/bin_fcst_obs.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Create a bi-variate histogram of forecast, observation pairs — bin_fcst_obs","text":"harp_verif list.","code":""},{"path":[]},{"path":"/reference/bind_bootstrap_score.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Bind output of bootstrap_score / pooled_bootstrap_score — bind_bootstrap_score","text":"","code":"bind_bootstrap_score(...)"},{"path":"/reference/bind_bootstrap_score.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Bind output of bootstrap_score / pooled_bootstrap_score — bind_bootstrap_score","text":"... list outputs bootstrap_score() / pooled_bootstrap_score()","code":""},{"path":"/reference/bind_bootstrap_score.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Bind output of bootstrap_score / pooled_bootstrap_score — bind_bootstrap_score","text":"object class harp_bootstrap","code":""},{"path":"/reference/bind_point_verif.html","id":null,"dir":"Reference","previous_headings":"","what":"Bind harp verification objects into a single object — bind_point_verif","title":"Bind harp verification objects into a single object — bind_point_verif","text":"plotting may desirable combine various different verifications single object. example, verification different parameters set forecast models.","code":""},{"path":"/reference/bind_point_verif.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Bind harp verification objects into a single object — bind_point_verif","text":"","code":"bind_point_verif(...)"},{"path":"/reference/bind_point_verif.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Bind harp verification objects into a single object — bind_point_verif","text":"... Verification objects combined. Can individual objects list.","code":""},{"path":"/reference/bind_point_verif.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Bind harp verification objects into a single object — bind_point_verif","text":"harp verification object attributes moved columns combined new attributes.","code":""},{"path":"/reference/bootstrap_score.html","id":null,"dir":"Reference","previous_headings":"","what":"Bootstrap a score — bootstrap_score","title":"Bootstrap a score — bootstrap_score","text":"Compute confidence intervals verification scores using bootstrap method. one forecast confidence differences forecasts also computed.","code":""},{"path":"/reference/bootstrap_score.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Bootstrap a score — bootstrap_score","text":"","code":"bootstrap_score( .fcst, score_function, parameter, n, groupings = \"leadtime\", confidence_interval = 0.95, ... )"},{"path":"/reference/bootstrap_score.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Bootstrap a score — bootstrap_score","text":".fcst harp_fcst object. score_function name verification function bootstrap . parameter parameter gives name observations columns. n Th number bootstrap replicants. groupings groups compute verification scores . default leadtime. ... arguments score function. e.g. thresholds.","code":""},{"path":"/reference/bootstrap_verify.html","id":null,"dir":"Reference","previous_headings":"","what":"Bootstrap a verification function — bootstrap_verify","title":"Bootstrap a verification function — bootstrap_verify","text":"bootstrap_verify used compute verification scores confidence intervals. one fcst_model exists input harp_list object, statistical significance differences verification scores fcst_models computed. statistical testing done using bootstrap method whereby scores computed repeatedly random samples input data.","code":""},{"path":"/reference/bootstrap_verify.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Bootstrap a verification function — bootstrap_verify","text":"","code":"bootstrap_verify( .fcst, verif_func, obs_col, n, groupings = \"lead_time\", pool_by = NULL, conf = 0.95, min_cases = 4, perfect_scores = perfect_score(), parallel = FALSE, num_cores = NULL, show_progress = TRUE, ... )"},{"path":"/reference/bootstrap_verify.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Bootstrap a verification function — bootstrap_verify","text":".fcst harp_list object column observations. verif_func harpPoint verification function bootstrap. obs_col observations column harp_list object. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}} n number bootstrap replicates. groupings groups compute scores. See group_by information grouping works. pool_by block bootstrap, quoted column name use pool data blocks. overlapping blocks data frame column common harp_list object input column named \"pool\" pool data belong . See Details. conf confidence interval compute. min_cases minimum number cases required group. block bootstrapping minimum number blocks. perfect_scores values score perfect score. parallel Set TRUE use parallel processing bootstrapping. Requires parallel package. num_cores parallel = TRUE, number cores use parallel processing. NULL, number cores detected parallel::detectCores() used. show_progress Logical. Set TRUE show progress bar. feature available parallel = TRUE ... arguments verif_func","code":""},{"path":"/reference/bootstrap_verify.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Bootstrap a verification function — bootstrap_verify","text":"harp_point_verif object extra columns upper lower confidence bounds scores percent replicates \"better\" one fcst_models input harp_list_object","code":""},{"path":"/reference/bootstrap_verify.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Bootstrap a verification function — bootstrap_verify","text":"data auto-correlated block bootstrap may used, whereby data pooled groups serial dependencies maintained. Rather sampling individual data points randomly, pools data points sampled randomly. pools taken column passed pool_by argument. use overlapping block bootstrap data frame passed pool_by, one column common harp_list object input column named \"pool\" labels pool row . ensures correct number overlapping pools used bootstrap replicate. make_bootstrap_pools can used get data frame overlapping pools. Bootstrapping can quite slow since many replicates computed. order speed process , bootstrap_verify also works parallel whereby replicates computed individual cores parallel rather serial. can achieved setting parallel = TRUE. default behaviour use cores, number cores can set num_cores argument.","code":""},{"path":"/reference/check_obs_against_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Observation error check against forecast — check_obs_against_fcst","title":"Observation error check against forecast — check_obs_against_fcst","text":"stratification standard deviation forecast forecast cycles models (mname), case eps forecast, ensemble members computed. difference observation forecast expected smaller number multiples standard deviation. number multiples standard deviation can supplied default value used depending parameter.","code":""},{"path":"/reference/check_obs_against_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Observation error check against forecast — check_obs_against_fcst","text":"","code":"check_obs_against_fcst( .fcst, parameter, num_sd_allowed = NULL, stratification = c(\"SID\", \"quarter_day\") )"},{"path":"/reference/check_obs_against_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Observation error check against forecast — check_obs_against_fcst","text":".fcst harp_df data frame, harp_list, observations column. parameter observations column. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. num_sd_allowed number standard deviations forecast difference forecast observation must smaller . stratification columns stratify data computing allowed tolerance. cases column must exist input data, \"quarter_day\" can passed divide observations classes [0, 6), [6, 12), [12, 18) [18, 24) hour day. default behaviour stratify station (\"SID\") \"quarter_day\".","code":""},{"path":"/reference/check_obs_against_fcst.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Observation error check against forecast — check_obs_against_fcst","text":"object sames class .fcst attribute named removed_cases containing data frame removed cases.","code":""},{"path":"/reference/det_probabilities.html","id":null,"dir":"Reference","previous_headings":"","what":"Compute binary probabilities for deterministic forecasts — det_probabilities","title":"Compute binary probabilities for deterministic forecasts — det_probabilities","text":"Compute binary probabilities deterministic forecasts","code":""},{"path":"/reference/det_probabilities.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compute binary probabilities for deterministic forecasts — det_probabilities","text":"","code":"det_probabilities(.fcst, parameter, thresholds, obs_probabilities = TRUE)"},{"path":"/reference/det_probabilities.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compute binary probabilities for deterministic forecasts — det_probabilities","text":".fcst harp_list object harp_det_point_df deta frames, harp_det_point_df data frame. parameter name column observed data. thresholds numeric vector thresholds compute probabilities. obs_probabilities logical indicating whether compute binary probabilities observations.","code":""},{"path":"/reference/det_probabilities.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compute binary probabilities for deterministic forecasts — det_probabilities","text":"object class .fcst columns threshold, fcst_prob optionally obs_prob instead raw forecast column.","code":""},{"path":"/reference/det_verify.html","id":null,"dir":"Reference","previous_headings":"","what":"Compute verification scores for deterministic forecasts. — det_verify","title":"Compute verification scores for deterministic forecasts. — det_verify","text":"Compute verification scores deterministic forecasts.","code":""},{"path":"/reference/det_verify.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compute verification scores for deterministic forecasts. — det_verify","text":"","code":"det_verify( .fcst, parameter, thresholds = NULL, groupings = \"lead_time\", circle = NULL, hexbin = TRUE, num_bins = 30, show_progress = TRUE, ... ) # S3 method for harp_det_point_df det_verify( .fcst, parameter, thresholds = NULL, groupings = \"lead_time\", circle = NULL, hexbin = TRUE, num_bins = 30, show_progress = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/det_verify.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compute verification scores for deterministic forecasts. — det_verify","text":".fcst harp_df harp_list object tables column observations, single forecast table. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. thresholds numeric vector thresholds compute threshold based scores. Set NULL (default) compute summary scores. groupings groups compute scores. See group_by information grouping works. circle set parameter assumed cyclic bias calculations. distance around circle units parameter, typically value 360 degrees 2 * pi radians. hexbin Logical. Whether compute hexbins forecast, observation pairs. Defaults TRUE. See bin_fcst_obs details. num_bins number bins partition observations hexbin computation. show_progress Logical - whether show progress bars. Defaults TRUE. ... Reserved methods. fcst_model name forecast model use fcst_model column output. function dispatched harp_list object, names harp_list automatically used.","code":""},{"path":"/reference/dplyr_list.html","id":null,"dir":"Reference","previous_headings":"","what":"dplyr verbs for lists — dplyr_list","title":"dplyr verbs for lists — dplyr_list","text":"list data frames, output verification function, may want wrangle data data frames time. can achieved using dplyr verb followed _list. data frames function applicable modified data frame returned. verb fails (e.g. specified columns exist), data frame silently returned unmodified","code":""},{"path":"/reference/dplyr_list.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"dplyr verbs for lists — dplyr_list","text":"","code":"mutate_list(.list, ...) filter_list(.list, ...) select_list(.list, ...)"},{"path":"/reference/dplyr_list.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"dplyr verbs for lists — dplyr_list","text":".list list data frames ... arguments dplyr verb","code":""},{"path":"/reference/dplyr_list.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"dplyr verbs for lists — dplyr_list","text":"list attrbutes input .list","code":""},{"path":[]},{"path":"/reference/ecoval.html","id":null,"dir":"Reference","previous_headings":"","what":"Compute economic value score — ecoval","title":"Compute economic value score — ecoval","text":"Compute economic value score","code":""},{"path":"/reference/ecoval.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compute economic value score — ecoval","text":"","code":"ecoval(obs, pred, costloss, thresholds)"},{"path":"/reference/ecoval.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compute economic value score — ecoval","text":"obs vector observations (value 0,1) pred vector probabilities 0,1. costloss vector cost/loss ratios thresholds vector threshold probabilities","code":""},{"path":"/reference/ecoval.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compute economic value score — ecoval","text":"list cl, value, Vmax, Venv, H, F, s, n","code":""},{"path":"/reference/ens_brier.html","id":null,"dir":"Reference","previous_headings":"","what":"Brier score and its decomposition for an ensemble. — ens_brier","title":"Brier score and its decomposition for an ensemble. — ens_brier","text":"Brier score decomposition ensemble.","code":""},{"path":"/reference/ens_brier.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Brier score and its decomposition for an ensemble. — ens_brier","text":"","code":"ens_brier( .fcst, parameter, thresholds, groupings = \"lead_time\", climatology = \"sample\", rel_probs = NA, num_ref_members = NA, keep_score = c(\"both\", \"brier\", \"reliability\"), show_progress = TRUE, ... ) # S3 method for harp_ens_point_df ens_brier( .fcst, parameter, thresholds, groupings = \"lead_time\", climatology = \"sample\", rel_probs = NA, num_ref_members = NA, keep_score = \"both\", show_progress = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/ens_brier.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Brier score and its decomposition for an ensemble. — ens_brier","text":".fcst harp_df harp_list object tables column observations, single forecast table. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. thresholds numeric vector thresholds compute threshold based scores. Set NULL (default) compute summary scores. groupings groups compute scores. See group_by information grouping works. climatology climatology use Brier Skill Score. Can \"sample\" sample climatology (default), named list elements eps_model member use member eps model harp_fcst object climatology, data frame columns threshold climatology also optionally lead_time. rel_probs Probabilities use reliability diagrams. Set NA (default) select automatically. num_ref_members \"fair\" scores, score scaled valid number ensemble members. Set NA (default) modify score. show_progress Logical - whether show progress bars. Defaults TRUE. ... Reserved methods. fcst_model name forecast model use fcst_model column output. function dispatched harp_list object, names harp_list automatically used.","code":""},{"path":"/reference/ens_brier.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Brier score and its decomposition for an ensemble. — ens_brier","text":"data frame data grouped groupings column(s) columns brier_score, brier_skill_score deomposition brier score - brier_score_reliability, brier_score_resolution brier_score_uncertainty.","code":""},{"path":"/reference/ens_crps.html","id":null,"dir":"Reference","previous_headings":"","what":"Continuous Rank Probability Score (CRPS) for an ensemble. — ens_crps","title":"Continuous Rank Probability Score (CRPS) for an ensemble. — ens_crps","text":"CRPS decomposition computed columns harp_list, harp_ens_grid_df object. Typically scores aggregated lead time, grouping variables cam chosen.","code":""},{"path":"/reference/ens_crps.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Continuous Rank Probability Score (CRPS) for an ensemble. — ens_crps","text":"","code":"ens_crps( .fcst, parameter, groupings = \"lead_time\", num_ref_members = NA, keep_full_output = FALSE, show_progress = TRUE, ... ) # S3 method for harp_ens_point_df ens_crps( .fcst, parameter, groupings = \"lead_time\", num_ref_members = NA, keep_full_output = FALSE, show_progress = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/ens_crps.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Continuous Rank Probability Score (CRPS) for an ensemble. — ens_crps","text":".fcst harp_df harp_list object tables column observations, single forecast table. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. groupings groups compute scores. See group_by information grouping works. num_ref_members \"fair\" scores, score scaled valid number ensemble members. Set NA (default) modify score. show_progress Logical - whether show progress bars. Defaults TRUE. ... Reserved methods. fcst_model name forecast model use fcst_model column output. function dispatched harp_list object, names harp_list automatically used.","code":""},{"path":"/reference/ens_crps.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Continuous Rank Probability Score (CRPS) for an ensemble. — ens_crps","text":"object format inputs data grouped groupings column(s) columns crps, crps_pot crps_rel.","code":""},{"path":"/reference/ens_probabilities.html","id":null,"dir":"Reference","previous_headings":"","what":"Compute probabilities of threshold exceedence for ensemble forecasts — ens_probabilities","title":"Compute probabilities of threshold exceedence for ensemble forecasts — ens_probabilities","text":"Compute probabilities threshold exceedence ensemble forecasts","code":""},{"path":"/reference/ens_probabilities.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compute probabilities of threshold exceedence for ensemble forecasts — ens_probabilities","text":"","code":"ens_probabilities(.fcst, thresholds, parameter = NULL)"},{"path":"/reference/ens_probabilities.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compute probabilities of threshold exceedence for ensemble forecasts — ens_probabilities","text":".fcst harp_df harp_list object tables column observations, single forecast table. thresholds numeric vector thresholds compute threshold based scores. Set NULL (default) compute summary scores. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}.","code":""},{"path":"/reference/ens_probabilities.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compute probabilities of threshold exceedence for ensemble forecasts — ens_probabilities","text":"harp_list object data frame columns threshold, fcst_prob obs_prob instead columns member forecast.","code":""},{"path":"/reference/ens_rank_histogram.html","id":null,"dir":"Reference","previous_headings":"","what":"Rank histogram for an ensemble. — ens_rank_histogram","title":"Rank histogram for an ensemble. — ens_rank_histogram","text":"rank histogram computed columns harp_list object. Typically scores aggregated lead time, grouping variables can chosen.","code":""},{"path":"/reference/ens_rank_histogram.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Rank histogram for an ensemble. — ens_rank_histogram","text":"","code":"ens_rank_histogram( .fcst, parameter, groupings = \"lead_time\", jitter_fcst = NULL, show_progress = TRUE, ... ) # S3 method for harp_ens_point_df ens_rank_histogram( .fcst, parameter, groupings = \"lead_time\", jitter_fcst = NULL, show_progress = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/ens_rank_histogram.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Rank histogram for an ensemble. — ens_rank_histogram","text":".fcst harp_df harp_list object tables column observations, single forecast table. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. groupings groups compute scores. See group_by information grouping works. jitter_fcst function perturb forecast values . used account observation error rank histogram. statistics likely make little difference since expected observations mean error zero. show_progress Logical - whether show progress bars. Defaults TRUE. ... Reserved methods. fcst_model name forecast model use fcst_model column output. function dispatched harp_list object, names harp_list automatically used.","code":""},{"path":"/reference/ens_rank_histogram.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Rank histogram for an ensemble. — ens_rank_histogram","text":"object format inputs data grouped groupings column(s) columns rank rank_count nested together column name rank_histogram.","code":""},{"path":"/reference/ens_read_and_verify.html","id":null,"dir":"Reference","previous_headings":"","what":"Read forecast and observations and verify. — ens_read_and_verify","title":"Read forecast and observations and verify. — ens_read_and_verify","text":"function likely replaced future harp versions since rather cumbersome. thought still needs go finding usable API. wrapper verification process. Forecasts observations read , filtered common cases, errors checked, full verification done scores. minimise memory usage, verification can done one lead time time. also possible parallelise process using example mclapply, future_map.","code":""},{"path":"/reference/ens_read_and_verify.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Read forecast and observations and verify. — ens_read_and_verify","text":"","code":"ens_read_and_verify( start_date, end_date, parameter, fcst_model, fcst_path, obs_path, lead_time = seq(0, 48, 3), num_iterations = length(lead_time), verify_members = TRUE, thresholds = NULL, members = NULL, vertical_coordinate = c(NA_character_, \"pressure\", \"model\", \"height\"), fctable_file_template = \"fctable\", obsfile_template = \"obstable\", groupings = \"lead_time\", by = \"6h\", lags = \"0s\", merge_lags_on_read = TRUE, lag_fcst_models = NULL, parent_cycles = NULL, lag_direction = 1, fcst_shifts = NULL, keep_unshifted = FALSE, drop_neg_leadtimes = TRUE, climatology = \"sample\", stations = NULL, scale_fcst = NULL, scale_obs = NULL, spread_drop_member = NULL, jitter_fcst = NULL, common_cases_only = TRUE, common_cases_xtra_cols = NULL, check_obs_fcst = TRUE, gross_error_check = TRUE, min_allowed = NULL, max_allowed = NULL, num_sd_allowed = NULL, show_progress = FALSE, verif_path = NULL )"},{"path":"/reference/ens_read_and_verify.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Read forecast and observations and verify. — ens_read_and_verify","text":"start_date Start date verification. numeric character. YYYYMMDD(HH)(mm). end_date End date verification. numeric character. parameter parameter verify. fcst_model forecast model(s) verify. Can single string character vector model names. fcst_path path forecast FCTABLE files. obs_path path observation OBSTABLE files. lead_time lead times verify. num_iterations number iterations per verification calculation. default number iterations lead times. small number iterations set, may useful set show_progress = TRUE. higher number iterations, smaller amount data held memory one time. verify_members Whether verify individual members ensemble. Even thresholds supplied, summary scores computed. wish compute categorical scores, separate det_verify function must used. thresholds thresholds compute categorical scores . members members retrieve reading EPS forecast. select members forecast models, numeric vector. specific members specific models named list element name forecast model containing numeric vector. e.g. members = list(eps_model1 = seq(0, 3), eps_model2 = c(2, 3)). multi model ensembles, element named list contain another named list sub model name followed desired members, e.g. members = list(eps_model1 = list(sub_model1 = seq(0, 3), sub_model2 = c(2, 3))) vertical_coordinate vertical co-ordinate. fctable_file_template template file names files read . normally one \"fctable_*\" templates can seen show_file_templates. Can single string, character vector list length fcst_model. named, order templates assumed fcst_model. named, names must match entries fcst_model. obsfile_template template OBSTABLE files - default \"obstable\", OBSTABLE_{YYYY}.sqlite. groupings groups verify . default \"leadtime\". Another common grouping might groupings = c(\"leadtime\", \"fcst_cycle\"). frequency forecast cycles verify. lags lagged forecasts, lags passed read_point_forecast(). merge_lags_on_read Logical. Whether merge lagged ensemble members ensemble. default behaviour. FALSE, lag_forecast used lagging. lag_fcst_models merge_lags_on_read = FALSE, names fcst_models lags applied. parent_cycles merge_lags_on_read = FALSE, parent cycles lagged forecasts. lag_direction merge_lags_on_read = FALSE, direction lagging. 1 Lags backwards time parent cycles -1 lags forwards time. fcst_shifts, keep_unshifted See shift_forecast. drop_neg_leadtimes Logical. Whether drop negative lead times may arise shifting. climatology climatology use Brier Skill Score. Can \"sample\" sample climatology (default), named list elements eps_model member use member eps model harp_fcst object climatology, data frame columns threshold climatology also optionally leadtime. stations stations verify . default use stations station_list common fcst_model domains. scale_fcst named list arguments scale_point_forecast. scale_obs names list arguments scale_point_obs. spread_drop_member members drop calculation ensemble variance standard deviation. harp_fcst objects, can numeric scalar - case recycled forecast models; list numeric vector length harp_fcst object, named list names corresponding names harp_fcst object. jitter_fcst function perturb forecast values . used account observation error rank histogram. statistics likely make little difference since expected observations mean error zero. common_cases_only Logical. Whether select common cases computing verification scores. default TRUE. common_cases_xtra_cols Extra columns use call common_cases check_obs_fcst Logical. Whether check errors observations comparing forecast values. gross_error_check Logical whether perform gross error check. min_allowed minimum value observation allow gross error check. set NULL default value parameter used. max_allowed maximum value observation allow gross error check. set NULL default value parameter used. num_sd_allowed number standard deviations forecast obseravtions within. Set NULL automotic value depeninding parameter. show_progress Logical - whether show progress bar. Defaults FALSE. verif_path set, verification files saved path.","code":""},{"path":"/reference/ens_read_and_verify.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Read forecast and observations and verify. — ens_read_and_verify","text":"list containing two data frames: ens_summary_scores ens_threshold_scores.","code":""},{"path":"/reference/ens_reliability.html","id":null,"dir":"Reference","previous_headings":"","what":"Reliability for an ensemble. — ens_reliability","title":"Reliability for an ensemble. — ens_reliability","text":"Reliability ensemble.","code":""},{"path":"/reference/ens_reliability.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Reliability for an ensemble. — ens_reliability","text":"","code":"ens_reliability( .fcst, parameter, thresholds, groupings = \"lead_time\", climatology = \"sample\", rel_probs = NA, show_progress = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/ens_reliability.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Reliability for an ensemble. — ens_reliability","text":".fcst harp_df harp_list object tables column observations, single forecast table. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. thresholds numeric vector thresholds compute threshold based scores. Set NULL (default) compute summary scores. groupings groups compute scores. See group_by information grouping works. climatology climatology use Brier Skill Score. Can \"sample\" sample climatology (default), named list elements eps_model member use member eps model harp_fcst object climatology, data frame columns threshold climatology also optionally lead_time. rel_probs Probabilities use reliability diagrams. Set NA (default) select automatically. show_progress Logical - whether show progress bars. Defaults TRUE. fcst_model name forecast model use fcst_model column output. function dispatched harp_list object, names harp_list automatically used. ... Reserved methods.","code":""},{"path":"/reference/ens_roc.html","id":null,"dir":"Reference","previous_headings":"","what":"Relative Operating Characteristics for an ensemble. — ens_roc","title":"Relative Operating Characteristics for an ensemble. — ens_roc","text":"Relative Operating Characteristics ensemble.","code":""},{"path":"/reference/ens_roc.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Relative Operating Characteristics for an ensemble. — ens_roc","text":"","code":"ens_roc( .fcst, parameter, thresholds, groupings = \"lead_time\", show_progress = TRUE, ... ) # S3 method for harp_ens_point_df ens_roc( .fcst, parameter, thresholds, groupings = \"lead_time\", show_progress = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/ens_roc.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Relative Operating Characteristics for an ensemble. — ens_roc","text":".fcst harp_df harp_list object tables column observations, single forecast table. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. thresholds numeric vector thresholds compute threshold based scores. Set NULL (default) compute summary scores. groupings groups compute scores. See group_by information grouping works. show_progress Logical - whether show progress bars. Defaults TRUE. ... Reserved methods. fcst_model name forecast model use fcst_model column output. function dispatched harp_list object, names harp_list automatically used.","code":""},{"path":"/reference/ens_roc.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Relative Operating Characteristics for an ensemble. — ens_roc","text":"data frame data grouped groupings column(s) nested column ROC row containing data frame columns: prob forecast probability bin, HR hit rate FAR false alarm rate. Use unnest unnest nested column.","code":""},{"path":"/reference/ens_spread_and_skill.html","id":null,"dir":"Reference","previous_headings":"","what":"Compute the skill (RMSE) and spread of an ensemble forecast — ens_spread_and_skill","title":"Compute the skill (RMSE) and spread of an ensemble forecast — ens_spread_and_skill","text":"ensemble mean spread computed columns harp_list object. Typically scores aggregated lead time grouping variables cam chosen. mean bias also computed.","code":""},{"path":"/reference/ens_spread_and_skill.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compute the skill (RMSE) and spread of an ensemble forecast — ens_spread_and_skill","text":"","code":"ens_spread_and_skill( .fcst, parameter, groupings = \"lead_time\", circle = NULL, spread_drop_member = NULL, jitter_fcst = NULL, show_progress = TRUE, ... ) # S3 method for harp_ens_point_df ens_spread_and_skill( .fcst, parameter, groupings = \"lead_time\", circle = NULL, spread_drop_member = NULL, jitter_fcst = NULL, show_progress = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/ens_spread_and_skill.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compute the skill (RMSE) and spread of an ensemble forecast — ens_spread_and_skill","text":".fcst harp_df harp_list object tables column observations, single forecast table. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. groupings groups compute scores. See group_by information grouping works. circle set parameter assumed cyclic bias calculations. distance around circle units parameter, typically value 360 degrees 2 * pi radians. spread_drop_member members drop calculation ensemble variance standard deviation. harp_fcst objects, can numeric scalar - case recycled forecast models; list numeric vector length harp_fcst object, named list names corresponding names harp_fcst object. jitter_fcst function perturb forecast values . used account observation error rank histogram. statistics likely make little difference since expected observations mean error zero. show_progress Logical - whether show progress bars. Defaults TRUE. ... Reserved methods. fcst_model name forecast model use fcst_model column output. function dispatched harp_list object, names harp_list automatically used.","code":""},{"path":"/reference/ens_spread_and_skill.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compute the skill (RMSE) and spread of an ensemble forecast — ens_spread_and_skill","text":"object format inputs data grouped groupings column(s) columns rmse, spread mean_bias.","code":""},{"path":"/reference/ens_value.html","id":null,"dir":"Reference","previous_headings":"","what":"Economic value for an ensemble. — ens_value","title":"Economic value for an ensemble. — ens_value","text":"Economic value ensemble.","code":""},{"path":"/reference/ens_value.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Economic value for an ensemble. — ens_value","text":"","code":"ens_value( .fcst, parameter, thresholds, groupings = \"lead_time\", show_progress = TRUE, ... ) # S3 method for harp_ens_point_df ens_value( .fcst, parameter, thresholds, groupings = \"lead_time\", show_progress = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/ens_value.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Economic value for an ensemble. — ens_value","text":".fcst harp_df harp_list object tables column observations, single forecast table. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. thresholds numeric vector thresholds compute threshold based scores. Set NULL (default) compute summary scores. groupings groups compute scores. See group_by information grouping works. show_progress Logical - whether show progress bars. Defaults TRUE. ... Reserved methods. fcst_model name forecast model use fcst_model column output. function dispatched harp_list object, names harp_list automatically used.","code":""},{"path":"/reference/ens_value.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Economic value for an ensemble. — ens_value","text":"data frame data grouped groupings column(s) nested column economic value row containing data frame columns: cl cost loss, value economic value. Use unnest unnest nested column.","code":""},{"path":"/reference/ens_verify.html","id":null,"dir":"Reference","previous_headings":"","what":"Compute all verification scores for an ensemble. — ens_verify","title":"Compute all verification scores for an ensemble. — ens_verify","text":"Compute verification scores ensemble.","code":""},{"path":"/reference/ens_verify.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compute all verification scores for an ensemble. — ens_verify","text":"","code":"ens_verify( .fcst, parameter, verify_members = TRUE, thresholds = NULL, groupings = \"lead_time\", circle = NULL, rel_probs = NA, num_ref_members = NA, spread_drop_member = NULL, jitter_fcst = NULL, climatology = \"sample\", hexbin = TRUE, num_bins = 30, rank_hist = TRUE, crps = TRUE, brier = TRUE, roc = TRUE, econ_val = TRUE, show_progress = TRUE, ... ) # S3 method for harp_ens_point_df ens_verify( .fcst, parameter, verify_members = TRUE, thresholds = NULL, groupings = \"lead_time\", circle = NULL, rel_probs = NA, num_ref_members = NA, spread_drop_member = NULL, jitter_fcst = NULL, climatology = \"sample\", hexbin = TRUE, num_bins = 30, rank_hist = TRUE, crps = TRUE, show_progress = TRUE, brier = TRUE, roc = TRUE, econ_val = TRUE, fcst_model = NULL, ... )"},{"path":"/reference/ens_verify.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compute all verification scores for an ensemble. — ens_verify","text":".fcst harp_df harp_list object tables column observations, single forecast table. parameter name column observations data. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. verify_members Whether verify individual members ensemble. Even thresholds supplied, summary scores computed. wish compute categorical scores, separate det_verify function must used. thresholds numeric vector thresholds compute threshold based scores. Set NULL (default) compute summary scores. groupings groups compute scores. See group_by information grouping works. circle set parameter assumed cyclic bias calculations. distance around circle units parameter, typically value 360 degrees 2 * pi radians. rel_probs Probabilities use reliability diagrams. Set NA (default) select automatically. num_ref_members \"fair\" scores, score scaled valid number ensemble members. Set NA (default) modify score. spread_drop_member members drop calculation ensemble variance standard deviation. harp_fcst objects, can numeric scalar - case recycled forecast models; list numeric vector length harp_fcst object, named list names corresponding names harp_fcst object. jitter_fcst function perturb forecast values . used account observation error rank histogram. statistics likely make little difference since expected observations mean error zero. climatology climatology use Brier Skill Score. Can \"sample\" sample climatology (default), named list elements eps_model member use member eps model harp_fcst object climatology, data frame columns threshold climatology also optionally lead_time. hexbin Logical. Whether compute hexbins forecast, observation pairs. Defaults TRUE. See bin_fcst_obs details. num_bins number bins partition observations hexbin computation. rank_hist Logical. Whether compute rank histogram. Defaults TRUE. Note computation rank histogram can slow large number (> 1000) groups. crps Logical. Whether compute CRPS. Defaults TRUE. brier Logical. Whether compute Brier score. Defaults TRUE. ignored thresholds set. roc Logical. Whether compute Relative Operating Characteristic (ROC). Defaults TRUE. ignored thresholds set. econ_val Logical. Whether compute economic value. Defaults TRUE. ignored thresholds set. show_progress Logical - whether show progress bars. Defaults TRUE. ... Reserved methods. fcst_model name forecast model use fcst_model column output. function dispatched harp_list object, names harp_list automatically used.","code":""},{"path":"/reference/ens_verify.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compute all verification scores for an ensemble. — ens_verify","text":"list containting three data frames: ens_summary_scores, ens_threshold_scores det_summary_scores.","code":""},{"path":"/reference/fcprob.html","id":null,"dir":"Reference","previous_headings":"","what":"Compute the forecast probabilities for given thresholds — fcprob","title":"Compute the forecast probabilities for given thresholds — fcprob","text":"Compute forecast probabilities given thresholds","code":""},{"path":"/reference/fcprob.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compute the forecast probabilities for given thresholds — fcprob","text":"","code":"fcprob(fc, thresholds)"},{"path":"/reference/fcprob.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compute the forecast probabilities for given thresholds — fcprob","text":"fc two dimensional array EPS data members columns. byrow thresholds compute probabilities ","code":""},{"path":"/reference/filter.harp_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Filter a harp_fcst object. — filter.harp_fcst","title":"Filter a harp_fcst object. — filter.harp_fcst","text":"Works table harp_fcst object way filter","code":""},{"path":"/reference/filter.harp_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Filter a harp_fcst object. — filter.harp_fcst","text":"","code":"# S3 method for harp_fcst filter(.fcst, ...)"},{"path":"/reference/filter.harp_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Filter a harp_fcst object. — filter.harp_fcst","text":".fcst harp_fcst object. ... Arguments filter","code":""},{"path":"/reference/first_validdate.html","id":null,"dir":"Reference","previous_headings":"","what":"Return the first valid date in a forecast — first_validdate","title":"Return the first valid date in a forecast — first_validdate","text":"version 0.1.0 reading data longer done start date end date, rather vector date-time strings making functions obsolete. unique_valid_dttm now appropriate function use. function intended used find first date fetch observations verify.","code":""},{"path":"/reference/first_validdate.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Return the first valid date in a forecast — first_validdate","text":"","code":"first_validdate(.fcst) last_validdate(.fcst)"},{"path":"/reference/first_validdate.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Return the first valid date in a forecast — first_validdate","text":".fcst harp_fcst object table containing column named validdate data unix time format.","code":""},{"path":"/reference/first_validdate.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Return the first valid date in a forecast — first_validdate","text":"first valid time YYYYMMDDhhmm format input object.","code":""},{"path":"/reference/gather_members.html","id":null,"dir":"Reference","previous_headings":"","what":"Convert EPS forecast data from wide format data frame to long format data\nframe. — gather_members","title":"Convert EPS forecast data from wide format data frame to long format data\nframe. — gather_members","text":"pivot_members now preferred method transforming long wide data frames since supports classes introduced version 0.1.0.","code":""},{"path":"/reference/gather_members.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Convert EPS forecast data from wide format data frame to long format data\nframe. — gather_members","text":"","code":"gather_members( .fcst, member_regex = \"_mbr[[:digit:]]+$|_mbr[[:digit:]]+_lag[[:graph:]]*$\" )"},{"path":"/reference/gather_members.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Convert EPS forecast data from wide format data frame to long format data\nframe. — gather_members","text":".fcst EPS forecast data frame wide format. member_regex Regular expression column names contain forecasts single member.","code":""},{"path":"/reference/gather_members.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Convert EPS forecast data from wide format data frame to long format data\nframe. — gather_members","text":"EPS data frame long format.","code":""},{"path":"/reference/harpPoint-package.html","id":null,"dir":"Reference","previous_headings":"","what":"harpPoint: Point verifition for NWP forecasts — harpPoint-package","title":"harpPoint: Point verifition for NWP forecasts — harpPoint-package","text":"Functions computing verification scores NWP forecasts. Part harp ecosystem.","code":""},{"path":[]},{"path":"/reference/harpPoint-package.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"harpPoint: Point verifition for NWP forecasts — harpPoint-package","text":"Maintainer: Andrew Singleton andrewts@met.Authors: Alex Deckmyn alex.deckmyn@meteo.","code":""},{"path":"/reference/jitter_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Jitter a forecast to account for observation errors. — jitter_fcst","title":"Jitter a forecast to account for observation errors. — jitter_fcst","text":"account observation errors ensemble verification forecast values can perturbed sampling specified error distribution. Jittering forecast likely effect ensemble spread rank histograms.","code":""},{"path":"/reference/jitter_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Jitter a forecast to account for observation errors. — jitter_fcst","text":"","code":"jitter_fcst(.fcst, jitter_func, obs_col = NULL, ...)"},{"path":"/reference/jitter_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Jitter a forecast to account for observation errors. — jitter_fcst","text":".fcst object class 'harp_fcst'. jitter_func function applied forecast values. obs_col observations column used jitter function. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}}. ... arguments jitter_func.","code":""},{"path":"/reference/jitter_fcst.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Jitter a forecast to account for observation errors. — jitter_fcst","text":"object class .fcst jittered forecast values.","code":""},{"path":"/reference/jitter_fcst.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Jitter a forecast to account for observation errors. — jitter_fcst","text":"Note jitter function function works vector forecast values one two arguments. first argument forecast data, second argument, observations.","code":""},{"path":"/reference/join_models.html","id":null,"dir":"Reference","previous_headings":"","what":"Join all models into a single ensemble. — join_models","title":"Join all models into a single ensemble. — join_models","text":"function useful finding common cases models.","code":""},{"path":"/reference/join_models.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Join all models into a single ensemble. — join_models","text":"","code":"join_models(.fcst, join_type = \"inner\", name = \"joined_models\", ...)"},{"path":"/reference/join_models.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Join all models into a single ensemble. — join_models","text":".fcst harp_list object multimodel data merged merge_multimodel. join_type type join perform. See join. name name resulting model. ... arguments join.","code":""},{"path":"/reference/join_models.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Join all models into a single ensemble. — join_models","text":"harp_df data frame.","code":""},{"path":"/reference/lag_forecast.html","id":null,"dir":"Reference","previous_headings":"","what":"Lags a forecast — lag_forecast","title":"Lags a forecast — lag_forecast","text":"Lagging done supplying parent forecast cycles. function work cycles belong parent return lagged forecast members child cycles appended parent cycle.","code":""},{"path":"/reference/lag_forecast.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Lags a forecast — lag_forecast","text":"","code":"lag_forecast(.fcst, fcst_model, parent_cycles, direction = 1)"},{"path":"/reference/lag_forecast.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Lags a forecast — lag_forecast","text":".fcst harp_fcst object. fcst_model name forecast model harp_fcst object lagged. Must quoted. parent_cycles numeric vector forecast cycles form parents lagging. Members parent cycles child cycles. dierction direction lagging . 1 (default) lags backwards time parent cycles -1 lags forwards time.","code":""},{"path":"/reference/lag_forecast.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Lags a forecast — lag_forecast","text":"harp_fcst object fcst_model now containing lagged forecast.","code":""},{"path":"/reference/make_bootstrap_pools.html","id":null,"dir":"Reference","previous_headings":"","what":"Make pools for block bootstrapping — make_bootstrap_pools","title":"Make pools for block bootstrapping — make_bootstrap_pools","text":"block bootstrap using bootstrap_verify, blocks can passed data frame \"pool\" column telling bootstrap_verify pool data blocks. make_bootstrap_pools function make data frame.","code":""},{"path":"/reference/make_bootstrap_pools.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Make pools for block bootstrapping — make_bootstrap_pools","text":"","code":"make_bootstrap_pools(.fcst, pool_col, pool_length, overlap = FALSE)"},{"path":"/reference/make_bootstrap_pools.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Make pools for block bootstrapping — make_bootstrap_pools","text":".fcst harp_fcst object pool_col column used define pools. Can column name, quoted, unquoted. variable embraced - .e. wrapped {{}} pool_length length pool. Numeric character unit qualifier pool_col date-time format. unit qualifier can : \"s\" = seconds, \"m\" = minutes, \"h\" = hours, \"d\" = days. overlap Logical. Whether pools overlap.","code":""},{"path":"/reference/make_bootstrap_pools.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Make pools for block bootstrapping — make_bootstrap_pools","text":"data frame columns pool_col \"pool\".","code":""},{"path":"/reference/make_bootstrap_pools.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"Make pools for block bootstrapping — make_bootstrap_pools","text":"Typically block bootstrapping used serial auto-correlations data. example auto-correlations suspected forecasts, pools defined fcdate column create blocks data auto-correlations maintained. Pools may set overlap, whereby new pool created beginning new value pool_col. length pool defined units used pool_col - pool_col date-time column, pool_length assumed hours, though units can set adding qualifier letter: \"s\" = seconds, \"m\" = minutes, \"h\" = hours, \"d\" = days.","code":""},{"path":"/reference/make_bootstrap_pools.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Make pools for block bootstrapping — make_bootstrap_pools","text":"","code":"make_bootstrap_pools(ens_point_df, lead_time, 2) #> # A tibble: 24 × 2 #> lead_time pool #> #> 1 0 1 #> 2 1 1 #> 3 2 2 #> 4 3 2 #> 5 4 3 #> 6 5 3 #> 7 6 4 #> 8 7 4 #> 9 8 5 #> 10 9 5 #> # ℹ 14 more rows make_bootstrap_pools(ens_point_df, lead_time, 2, overlap = TRUE) #> # A tibble: 46 × 2 #> lead_time pool #> #> 1 0 1 #> 2 1 1 #> 3 1 2 #> 4 2 2 #> 5 2 3 #> 6 3 3 #> 7 3 4 #> 8 4 4 #> 9 4 5 #> 10 5 5 #> # ℹ 36 more rows # pool_col as a variable my_col <- \"lead_time\" make_bootstrap_pools(ens_point_df, {{my_col}}, 2) #> # A tibble: 24 × 2 #> lead_time pool #> #> 1 0 1 #> 2 1 1 #> 3 2 2 #> 4 3 2 #> 5 4 3 #> 6 5 3 #> 7 6 4 #> 8 7 4 #> 9 8 5 #> 10 9 5 #> # ℹ 14 more rows"},{"path":"/reference/merge_multimodel.html","id":null,"dir":"Reference","previous_headings":"","what":"Merge sub models of a multimodel ensemble into a single ensemble — merge_multimodel","title":"Merge sub models of a multimodel ensemble into a single ensemble — merge_multimodel","text":"multimodel ensembles read sub models stored. whole ensemble needed, function merge sub models single ensemble. Note renaming members sub model names retained. multi model ensembles, input silently returned unaltered.","code":""},{"path":"/reference/merge_multimodel.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Merge sub models of a multimodel ensemble into a single ensemble — merge_multimodel","text":"","code":"merge_multimodel(.fcst, keep_sub_models = TRUE)"},{"path":"/reference/merge_multimodel.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Merge sub models of a multimodel ensemble into a single ensemble — merge_multimodel","text":".fcst object class harp_fcst read read_point_forecast keep_sub_models Set FALSE discard sub models separate elements harp_fcst list. default behaviour keep .","code":""},{"path":"/reference/merge_multimodel.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Merge sub models of a multimodel ensemble into a single ensemble — merge_multimodel","text":"harp_fcst object one layer - element table forecast data model.","code":""},{"path":"/reference/mutate.harp_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Mutate columns from tables in a harp_fcst object. — mutate.harp_fcst","title":"Mutate columns from tables in a harp_fcst object. — mutate.harp_fcst","text":"Works way mutate, except runs tables harp_fcst object.","code":""},{"path":"/reference/mutate.harp_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Mutate columns from tables in a harp_fcst object. — mutate.harp_fcst","text":"","code":"# S3 method for harp_fcst mutate(.fcst, ...)"},{"path":"/reference/mutate.harp_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Mutate columns from tables in a harp_fcst object. — mutate.harp_fcst","text":".fcst harp_fcst object ... Arguments mutate","code":""},{"path":"/reference/mutate_at.harp_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Mutate selected columns from tables in a harp_fcst object. — mutate_at.harp_fcst","title":"Mutate selected columns from tables in a harp_fcst object. — mutate_at.harp_fcst","text":"Works way mutate_at, except runs tables harp_fcst object.","code":""},{"path":"/reference/mutate_at.harp_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Mutate selected columns from tables in a harp_fcst object. — mutate_at.harp_fcst","text":"","code":"mutate_at.harp_fcst(.fcst, .mutate_vars, .mutate_funs, ...)"},{"path":"/reference/mutate_at.harp_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Mutate selected columns from tables in a harp_fcst object. — mutate_at.harp_fcst","text":".fcst harp_fcst object ... Arguments mutate","code":""},{"path":"/reference/perfect_score.html","id":null,"dir":"Reference","previous_headings":"","what":"Set or modify and output the perfect scores for different verificaiton\nmetrics — perfect_score","title":"Set or modify and output the perfect scores for different verificaiton\nmetrics — perfect_score","text":"function returns named vector giving perfect score verification scores computed harpPoint give single value. perfect values can modified specifying named arguments perfect scores new scores can set way.","code":""},{"path":"/reference/perfect_score.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Set or modify and output the perfect scores for different verificaiton\nmetrics — perfect_score","text":"","code":"perfect_score(...)"},{"path":"/reference/perfect_score.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Set or modify and output the perfect scores for different verificaiton\nmetrics — perfect_score","text":"... Named arguments give values perfect score","code":""},{"path":"/reference/perfect_score.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Set or modify and output the perfect scores for different verificaiton\nmetrics — perfect_score","text":"named vector","code":""},{"path":"/reference/perfect_score.html","id":"ref-examples","dir":"Reference","previous_headings":"","what":"Examples","title":"Set or modify and output the perfect scores for different verificaiton\nmetrics — perfect_score","text":"","code":"perfect_score() #> bias rmse mae #> 0e+00 0e+00 0e+00 #> stde threat_score hit_rate #> 0e+00 1e+00 1e+00 #> miss_rate false_alarm_rate false_alarm_ratio #> 0e+00 0e+00 0e+00 #> heidke_skill_score pierce_skill_score kuiper_skill_score #> 1e+00 1e+00 1e+00 #> percent_correct frequency_bias equitable_threat_score #> 1e+00 1e+00 1e+00 #> odds_ratio log_odds_ratio odds_ratio_skill_score #> 1e+06 1e+06 1e+00 #> extreme_dependency_score symmetric_eds extreme_dependency_index #> 1e+00 1e+00 1e+00 #> symmetric_edi mean_bias spread #> 1e+00 0e+00 1e+06 #> spread_skill_ratio crps crps_potential #> 1e+00 0e+00 0e+00 #> crps_reliability fair_brier_score fair_crps #> 0e+00 0e+00 0e+00 #> brier_score brier_skill_score brier_score_reliability #> 0e+00 1e+00 0e+00 #> brier_score_resolution roc_area #> 1e+00 1e+00 perfect_score(bss = 1) #> bias rmse mae #> 0e+00 0e+00 0e+00 #> stde threat_score hit_rate #> 0e+00 1e+00 1e+00 #> miss_rate false_alarm_rate false_alarm_ratio #> 0e+00 0e+00 0e+00 #> heidke_skill_score pierce_skill_score kuiper_skill_score #> 1e+00 1e+00 1e+00 #> percent_correct frequency_bias equitable_threat_score #> 1e+00 1e+00 1e+00 #> odds_ratio log_odds_ratio odds_ratio_skill_score #> 1e+06 1e+06 1e+00 #> extreme_dependency_score symmetric_eds extreme_dependency_index #> 1e+00 1e+00 1e+00 #> symmetric_edi mean_bias spread #> 1e+00 0e+00 1e+06 #> spread_skill_ratio crps crps_potential #> 1e+00 0e+00 0e+00 #> crps_reliability fair_brier_score fair_crps #> 0e+00 0e+00 0e+00 #> brier_score brier_skill_score brier_score_reliability #> 0e+00 1e+00 0e+00 #> brier_score_resolution roc_area bss #> 1e+00 1e+00 1e+00 perfect_score(bias = -1) #> bias rmse mae #> -1e+00 0e+00 0e+00 #> stde threat_score hit_rate #> 0e+00 1e+00 1e+00 #> miss_rate false_alarm_rate false_alarm_ratio #> 0e+00 0e+00 0e+00 #> heidke_skill_score pierce_skill_score kuiper_skill_score #> 1e+00 1e+00 1e+00 #> percent_correct frequency_bias equitable_threat_score #> 1e+00 1e+00 1e+00 #> odds_ratio log_odds_ratio odds_ratio_skill_score #> 1e+06 1e+06 1e+00 #> extreme_dependency_score symmetric_eds extreme_dependency_index #> 1e+00 1e+00 1e+00 #> symmetric_edi mean_bias spread #> 1e+00 0e+00 1e+06 #> spread_skill_ratio crps crps_potential #> 1e+00 0e+00 0e+00 #> crps_reliability fair_brier_score fair_crps #> 0e+00 0e+00 0e+00 #> brier_score brier_skill_score brier_score_reliability #> 0e+00 1e+00 0e+00 #> brier_score_resolution roc_area #> 1e+00 1e+00"},{"path":"/reference/pipe.html","id":null,"dir":"Reference","previous_headings":"","what":"Pipe operator — %>%","title":"Pipe operator — %>%","text":"See magrittr::%>% details.","code":""},{"path":"/reference/pipe.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Pipe operator — %>%","text":"","code":"lhs %>% rhs"},{"path":[]},{"path":"/reference/pooled_bootstrap_score.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Bootstrapping by pools — pooled_bootstrap_score","text":"","code":"pooled_bootstrap_score( .fcst, score_function, parameter, n, pooled_by = \"fcdate\", groupings = \"leadtime\", confidence_interval = 0.95, min_cases = 25, perfect_scores = perfect_score(), show_progress = TRUE, ... )"},{"path":"/reference/pull.harp_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Pull columns from tables in a harp_fcst object. — pull.harp_fcst","title":"Pull columns from tables in a harp_fcst object. — pull.harp_fcst","text":"Works way pull, except runs tables harp_fcst object. means common columns objects can safely pulled.","code":""},{"path":"/reference/pull.harp_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Pull columns from tables in a harp_fcst object. — pull.harp_fcst","text":"","code":"# S3 method for harp_fcst pull(.fcst, ...)"},{"path":"/reference/pull.harp_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Pull columns from tables in a harp_fcst object. — pull.harp_fcst","text":".fcst harp_fcst object ... Arguments pull","code":""},{"path":"/reference/pull_stations.html","id":null,"dir":"Reference","previous_headings":"","what":"Pull station IDs from forecast or observations — pull_stations","title":"Pull station IDs from forecast or observations — pull_stations","text":"Pull station IDs forecast observations","code":""},{"path":"/reference/pull_stations.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Pull station IDs from forecast or observations — pull_stations","text":"","code":"pull_stations(.fcst)"},{"path":"/reference/pull_stations.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Pull station IDs from forecast or observations — pull_stations","text":".fcst harp_fcst object data frame","code":""},{"path":"/reference/pull_stations.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Pull station IDs from forecast or observations — pull_stations","text":"vector station IDs","code":""},{"path":"/reference/qair2rh.html","id":null,"dir":"Reference","previous_headings":"","what":"qair2rh — qair2rh","title":"qair2rh — qair2rh","text":"Convert specific humidity relative humidity","code":""},{"path":"/reference/qair2rh.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"qair2rh — qair2rh","text":"","code":"qair2rh(qair, temp, press = 1013.25)"},{"path":"/reference/qair2rh.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"qair2rh — qair2rh","text":"qair specific humidity, dimensionless (e.g. kg/kg) ratio water mass / total air mass temp degrees C press pressure mb","code":""},{"path":"/reference/qair2rh.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"qair2rh — qair2rh","text":"rh relative humidity, ratio actual water mixing ratio saturation mixing ratio","code":""},{"path":"/reference/qair2rh.html","id":"details","dir":"Reference","previous_headings":"","what":"Details","title":"qair2rh — qair2rh","text":"converting specific humidity relative humidity NCEP surface flux data RH Bolton 1980 computation Equivalent Potential Temperature http://www.eol.ucar.edu/projects/ceop/dm/documents/refdata_report/eqns.html function lifetd data.atmosphere package https://github.com/PecanProject/pecan/blob/master/modules/data.atmosphere/R/metutils.R","code":""},{"path":"/reference/qair2rh.html","id":"author","dir":"Reference","previous_headings":"","what":"Author","title":"qair2rh — qair2rh","text":"David LeBauer","code":""},{"path":"/reference/rankHistogram.html","id":null,"dir":"Reference","previous_headings":"","what":"Compute the rank histogram for an EPS — rankHistogram","title":"Compute the rank histogram for an EPS — rankHistogram","text":"Compute rank histogram EPS","code":""},{"path":"/reference/rankHistogram.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compute the rank histogram for an EPS — rankHistogram","text":"","code":"rankHistogram(obs, fc)"},{"path":"/reference/rankHistogram.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compute the rank histogram for an EPS — rankHistogram","text":"obs vector observations. fc two dimensional array EPS data members columns.","code":""},{"path":"/reference/reexports.html","id":null,"dir":"Reference","previous_headings":"","what":"Objects exported from other packages — reexports","title":"Objects exported from other packages — reexports","text":"objects imported packages. Follow links see documentation. dplyr vars","code":""},{"path":"/reference/rename.harp_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Rename columns from tables in a harp_fcst object. — rename.harp_fcst","title":"Rename columns from tables in a harp_fcst object. — rename.harp_fcst","text":"Works way rename, except runs tables harp_fcst object. means common columns objects can safely renamed.","code":""},{"path":"/reference/rename.harp_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Rename columns from tables in a harp_fcst object. — rename.harp_fcst","text":"","code":"# S3 method for harp_fcst rename(.fcst, ...)"},{"path":"/reference/rename.harp_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Rename columns from tables in a harp_fcst object. — rename.harp_fcst","text":".fcst harp_fcst object ... Arguments rename","code":""},{"path":"/reference/rename_with.harp_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Rename multiple columns from tables in a harp_fcst object. — rename_with.harp_fcst","title":"Rename multiple columns from tables in a harp_fcst object. — rename_with.harp_fcst","text":"Works way rename_with, except runs tables harp_fcst object. means common columns objects can safely renamed.","code":""},{"path":"/reference/rename_with.harp_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Rename multiple columns from tables in a harp_fcst object. — rename_with.harp_fcst","text":"","code":"# S3 method for harp_fcst rename_with(.fcst, ...)"},{"path":"/reference/rename_with.harp_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Rename multiple columns from tables in a harp_fcst object. — rename_with.harp_fcst","text":".fcst harp_fcst object ... Arguments rename_with","code":""},{"path":"/reference/roc.html","id":null,"dir":"Reference","previous_headings":"","what":"Compute ROC and area under ROC — roc","title":"Compute ROC and area under ROC — roc","text":"Compute ROC area ROC","code":""},{"path":"/reference/roc.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Compute ROC and area under ROC — roc","text":"","code":"roc(obs, pred, thresholds)"},{"path":"/reference/roc.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Compute ROC and area under ROC — roc","text":"obs vector observations (value 0,1) pred vector probabilities 0,1. thresholds vector threshold probabilities","code":""},{"path":"/reference/roc.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Compute ROC and area under ROC — roc","text":"list area vectors thresholds, H, F","code":""},{"path":"/reference/scale_point_forecast.html","id":null,"dir":"Reference","previous_headings":"","what":"Scale forecast data — scale_point_forecast","title":"Scale forecast data — scale_point_forecast","text":"scale_param now preferred function since works point gridded data. wish scale forecast values, example temperature data Kelvin want degrees C, function can used scale data.","code":""},{"path":"/reference/scale_point_forecast.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Scale forecast data — scale_point_forecast","text":"","code":"scale_point_forecast( .fcst, scale_factor, new_units = NULL, multiplicative = FALSE )"},{"path":"/reference/scale_point_forecast.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Scale forecast data — scale_point_forecast","text":".fcst harp_fcst object read read_point_forecast harpIO. scale_factor scaling factor. new_units name new units - set NULL, name changed. multiplicative scaling done multiplicatively, .e. new forecast value old value * scale_factor, set multiplicative TRUE. default (multiplicative = FALSE) scaling additively, .e. new forecast value old value + scale_factor.","code":""},{"path":"/reference/scale_point_forecast.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Scale forecast data — scale_point_forecast","text":"harp_fcst object forecast values scaled scale_factor.","code":""},{"path":"/reference/scale_point_obs.html","id":null,"dir":"Reference","previous_headings":"","what":"Scale observations data — scale_point_obs","title":"Scale observations data — scale_point_obs","text":"scale_param now preferred function since works point gridded data. wish scale observations values, example temperature data Kelvin want degrees C, function can used scale data.","code":""},{"path":"/reference/scale_point_obs.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Scale observations data — scale_point_obs","text":"","code":"scale_point_obs( .obs, parameter, scale_factor, new_units = NULL, multiplicative = FALSE )"},{"path":"/reference/scale_point_obs.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Scale observations data — scale_point_obs","text":".obs data frame point observations read read read_point_obs harpIO. parameter column name data scaled. Must unquoted. scale_factor scaling factor. new_units name new units - set NULL, name changed. multiplicative scaling done multiplicatively, .e. new forecast value old value * scale_factor, set multiplicative TRUE. default (multiplicative = FALSE) scaling additively, .e. new forecast value old value + scale_factor.","code":""},{"path":"/reference/scale_point_obs.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Scale observations data — scale_point_obs","text":"observations data frame parameter column scaled scale_factor.","code":""},{"path":"/reference/select.harp_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Select columns from tables in a harp_fcst object. — select.harp_fcst","title":"Select columns from tables in a harp_fcst object. — select.harp_fcst","text":"Works way select, except runs tables harp_fcst object. means common columns objects can safely selected.","code":""},{"path":"/reference/select.harp_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Select columns from tables in a harp_fcst object. — select.harp_fcst","text":"","code":"# S3 method for harp_fcst select(.fcst, ...)"},{"path":"/reference/select.harp_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Select columns from tables in a harp_fcst object. — select.harp_fcst","text":".fcst harp_fcst object ... Arguments select","code":""},{"path":"/reference/shift_forecast.html","id":null,"dir":"Reference","previous_headings":"","what":"Shift a forecast — shift_forecast","title":"Shift a forecast — shift_forecast","text":"function used shift start times lead times forecasts simulate lagging.","code":""},{"path":"/reference/shift_forecast.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Shift a forecast — shift_forecast","text":"","code":"shift_forecast( .fcst, fcst_shifts, keep_unshifted = FALSE, drop_negative_lead_times = TRUE )"},{"path":"/reference/shift_forecast.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Shift a forecast — shift_forecast","text":".fcst harp_fcst object created read_point_forecast, data frame columns including 'fcdate' (seconds) 'leadtime' (hours). fcst_shifts named list names exist '.fcst' single numeric value apply forecast models. list, element must numeric length 1 1 shift can applied forecast. shifts specified hours. Postive values shift forecast start dates forward time reduce lead times corresponding amounts. Negative values opposite.","code":""},{"path":"/reference/shift_forecast.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Shift a forecast — shift_forecast","text":"object class '.fcst' forecast start times lead times shifted forecast models number hours given 'fcst_shifts'.","code":""},{"path":"/reference/sort_members.html","id":null,"dir":"Reference","previous_headings":"","what":"Sort a 2d array. — sort_members","title":"Sort a 2d array. — sort_members","text":"Sort 2d array.","code":""},{"path":"/reference/sort_members.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Sort a 2d array. — sort_members","text":"","code":"sort_members(x, byrow = TRUE)"},{"path":"/reference/sort_members.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Sort a 2d array. — sort_members","text":"x two dimensional numeric array. byrow Set true sort rows, false sort columns.","code":""},{"path":"/reference/spread_members.html","id":null,"dir":"Reference","previous_headings":"","what":"Convert EPS forecast data from long format data frame to wide format data\nframe. — spread_members","title":"Convert EPS forecast data from long format data frame to wide format data\nframe. — spread_members","text":"pivot_members now preferred method transforming long wide data frames since supports classes introduced version 0.1.0.","code":""},{"path":"/reference/spread_members.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Convert EPS forecast data from long format data frame to wide format data\nframe. — spread_members","text":"","code":"spread_members(.fcst, ...)"},{"path":"/reference/spread_members.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Convert EPS forecast data from long format data frame to wide format data\nframe. — spread_members","text":".fcst EPS forecast data frame long format.","code":""},{"path":"/reference/spread_members.html","id":"value","dir":"Reference","previous_headings":"","what":"Value","title":"Convert EPS forecast data from long format data frame to wide format data\nframe. — spread_members","text":"EPS data frame wide format.","code":""},{"path":"/reference/transmute.harp_fcst.html","id":null,"dir":"Reference","previous_headings":"","what":"Transmute columns from tables in a harp_fcst object. — transmute.harp_fcst","title":"Transmute columns from tables in a harp_fcst object. — transmute.harp_fcst","text":"Works way transmute, except runs tables harp_fcst object. means common columns objects can safely arranged.","code":""},{"path":"/reference/transmute.harp_fcst.html","id":"ref-usage","dir":"Reference","previous_headings":"","what":"Usage","title":"Transmute columns from tables in a harp_fcst object. — transmute.harp_fcst","text":"","code":"# S3 method for harp_fcst transmute(.fcst, ...)"},{"path":"/reference/transmute.harp_fcst.html","id":"arguments","dir":"Reference","previous_headings":"","what":"Arguments","title":"Transmute columns from tables in a harp_fcst object. — transmute.harp_fcst","text":".fcst harp_fcst object ... Arguments transmute","code":""},{"path":"/news/index.html","id":"harppoint-v020","dir":"Changelog","previous_headings":"","what":"harpPoint v0.2.0","title":"harpPoint v0.2.0","text":"major update. changes internal, changes may lead problems downstream scripts listed along selected new features.","code":""},{"path":"/news/index.html","id":"breaking-changes-0-2-0","dir":"Changelog","previous_headings":"","what":"Breaking changes","title":"harpPoint v0.2.0","text":"Default groupings argument changed \"leadtime\" \"lead_time\" consistency changes {harpIO}. Scripts use verification functions \"leadtime\", \"validdate\" \"fcdate\" values groupings changed \"lead_time\", \"valid_dttm\" \"fcst_dttm\" respectively. Verification outputs now use column name fcst_model instead mname consistency throughout harp. Attributes verification outputs changed include forecast dates, stations groupings used verification. scripts make use attributes updated reflect new attributes. scale_point_forecast() scale_point_obs() deprecated. scale_param() used instead. gather_members() spread_members() deprecated. pivot_members() used instead. first_validdate() last_validdate() deprecated. unique_valid_dttm() used instead. pull_stations() deprecated. unique_stations() used instead. bootstrap_score(), pooled_bootstrap_score() bind_bootstrap_score() defunct. bootstrap_verify() bind_point_verif() used instead.","code":""},{"path":"/news/index.html","id":"selected-new-features-0-2-0","dir":"Changelog","previous_headings":"","what":"Selected new features","title":"harpPoint v0.2.0","text":"Verification functions gained new progress bars generally less verbose , restricting messages progress computing scores different verification groups. New verification score hexbin. gives data frame essentially heat map forecast - observation value pairs. New class, attributes print method verification function outputs. jitter_fcst() now accepts vectorized functions lot faster.","code":""},{"path":"/news/index.html","id":"harppoint-v009","dir":"Changelog","previous_headings":"","what":"harpPoint v0.0.9","title":"harpPoint v0.0.9","text":"version basically unchanged since late 2021 / early 2022. officially tagged v0.0.9 November 2023","code":""}]
diff --git a/docs/sitemap.xml b/docs/sitemap.xml
index b4dc517..9d4dc76 100644
--- a/docs/sitemap.xml
+++ b/docs/sitemap.xml
@@ -15,6 +15,9 @@
/index.html
+
+ /news/index.html
+
/reference/arrange.harp_fcst.html