Skip to contents
-
add_horizons()
- Add columns indicating the horizon to forecasts or scores
-
assert_needed_env_vars()
- Assert that needed environment variables are set
-
benchmark_performance()
- Benchmark performance of wastewater and hospital admissions model foreast runs
-
check_any_flags()
- Check whether a model run had any flags.
-
clean_and_filter_nwss_data()
- Initial subsetting of NWSS data
-
clean_flag_df()
- Clean flags from a real-time forecast
-
clean_ww_data()
- Clean wastewater data
-
combine_and_summarize_ww_data()
- Combine and summarize wastewater data across all forecast dates and locations
-
combine_hub_and_local_wis()
- Combine the hosp only real-time scores and cfa real- time scores from github
-
combine_outputs()
- Combine outputs
-
compose_example_scores_fig()
- Compose figure showing examples of forecasts and their scores.
-
compose_hub_fig()
- Functions for creating multi-panel composite figures Make a figure showing comparisons between CFA models and other COVIDHub models
-
compose_pred_actual_fig()
- Compose figure showing predictions and observations for three example forecasts.
-
compose_rel_performance_fig()
- Make a multi-panel figure summarizing relative performance.
-
compute_flags()
- Compute diagnostic flags from raw stanfit objects
-
compute_forecast_difference()
- Compute the the posterior of the difference between model predictions.
-
create_hub_submissions()
- Create hub submission files
-
create_mock_submission_scores()
- Create mock submission scores
-
date_of_ww_data()
- Date of wastewater data
-
density_plot_std_rank()
- Make a figure of the distribution of standardized WIS rank
-
.relative_metric_display_name()
- Internal function for display names of relative metrics
-
eval_fit()
- Fit model for evaluation
-
eval_postprocess()
- Postprocess model for evaluation.
-
exclude_hosp_outliers()
- Exclude hospital outliers
-
fit_hosp_trend()
- Fit a hosp trend with BRMS
-
fit_recent_trends()
- Fit recent hospital admission and wastewater trends using BRMS
-
fit_ww_trend()
- Fit lab-site-level recent wastewater trends with BRMS.
-
flusight_location_table
- flusight_location_table dataset
-
forecast_interval_coverage_plot()
- Plot interval coverage at specified ranges
-
forecast_output_path()
- Standard path to an output directory for a specific forecast
-
forecast_qq_plot()
- Functions for plots that assess forecast calibration Get quantile-quantile plot
-
format_for_hub()
- Format quantiles for the current COVID Forecast Hub
-
format_scores_for_comparison()
- Format the hosp only real time scores for comparison to the other real time models
-
geom_decomposed_scores()
- Composite geometric object for barplots of scores decomposed into undeprediction, overprediction, and dispersion components
-
get_add_ww_metadata()
- Get additional wastewater metadata
-
get_box_plot()
- Get box plot by date and scenario
-
get_convergence_df()
- Get convergence dataframe
-
get_date_locs_excluded()
- Get forecast date locations that we excluded in the model run
-
get_diagnostic_flags()
- Get diagnostic flags
-
get_epidemic_phases_from_rt()
- Get epidemic phases from R(t)
-
get_filepath()
- Get filepath
-
get_growth_rate_plot()
- Compare growth rates in wastewater and hospital admissions data
-
get_heatmap_metadata()
- Get a heatmap of the metadata of reasons for excluding forecasts from analysis
-
get_heatmap_metadata_hub()
- Get a heatmap of the metadata of Hub models submitted
-
get_heatmap_scores()
- Get heatmap of scores by scenario
-
get_ind_forecast_score()
- Get an individual forecast score summary for a particular forecast date and location
-
get_input_hosp_data()
- Get input hospital admissions data
-
get_input_ww_data()
- Get input wastewater data
-
get_last_hosp_data_date()
- Get last hospital admissions data point date
-
get_last_hosp_data_date_map()
- Get a map of the location, forecast date, and last hospital admissions data date
-
get_model_draws_w_data()
- Get model draws combined with input and evaluation data
-
get_model_path()
- Get model path
-
get_n_states_improved_plot()
- Get bar chart of the number of states improved compared to without wastewater over each forecast date
-
get_object_loader()
- Generate a function for loading raw output objects for a location/forecast date/scenario trio.
-
get_object_saver()
- Generate a function for saving raw output objects for a location/forecast date/scenario trio.
-
get_plot_bias_over_time()
- Get a plot of bias over time
-
get_plot_hosp_data_comparison()
- Get plot of hospital admissions data compared to model draws
-
get_plot_quantile_comparison()
- Get plot quantile comparison
-
get_plot_raw_scores()
- Get plot raw scores
-
get_plot_score_by_horizon_t()
- Get plot of scores by horizon over time
-
get_plot_scores_and_forecasts()
- Got a plot of forecasts and scores for a single location and date
-
get_plot_scores_w_data()
- Get plot scores with evaluation data overlaid
-
get_plot_sites_vs_performance()
- Make a scatterplot comparing number of wastewater sites to model performance.
-
get_plot_summarized_scores()
- Get plot summarized scores
-
get_plot_theme()
- Get standardized plot theme to add to figures
-
get_plot_ww_data()
- Plot wastewater evaluation data
-
get_plot_ww_data_comparison()
- Get plot of wastewater data compared to model draws
-
get_raw_output_suffix()
- Generate a standard-format output suffix for saving raw output for a location/forecast date/scenario trio.
-
get_rel_wis_all_time()
- Get the relative wis from the hub formatted
-
get_rel_wis_real_time()
- Get the relative wis for the real time scores
-
get_scenario_site_ids()
- Get scenario site ids
-
get_score_summary_tables()
- Make a summary of the with and without wastewater comparison scores
-
get_state_level_quantiles()
- Get quantiles for state-level generated quantities
-
get_state_level_ww_quantiles()
- Get quantiles for site-lab level wastewater
-
get_stats_improved_forecasts()
- Get stats on number of improved forecasts
-
get_summary_metadata()
- Get a summary table of the number of forecasts excluded for each reason
-
get_summary_ww_table()
- Get wastewater data summary tables
-
get_table_sufficient_ww()
- Get table of location-forecast dates with sufficient wastewater
-
get_ww_data_flags()
- Get table of location-forecast dates with sufficient wastewater
-
heatmap_scores_by_loc_date()
- Plot a heatmap of avg forecast performance by location and forecast date
-
hub_average_score_table()
- Make summary table of WIS scores in Hub models overall
-
load_data_and_summarize()
- Load data and summarize metadata
-
load_real_time_forecast()
- Load a real-time forecast
-
load_real_time_quantile_fcsts()
- Load real-time quantile outputs
-
make_baseline_score_table()
- Make baseline score table
-
make_df()
- Make dataframe
-
make_fig2_ct_supp()
- Make concentration fit and forecast figure for supplement
-
multi_location_crps_figure()
- Generate bar plots of CRPS for each model in different locations
-
nhsn_soda_query()
- Construct a Socrata open data API (SODA) query for the NSHN dataset
-
order_col()
- Light wrapper function for converting a dataframe column to an ordered factors.
-
order_horizons()
- Get the order of the categorical horizons for plotting
-
order_locations()
- Get the order of the locations for plotting
-
order_periods()
- Get the order of the time periods for plotting
-
order_phases()
- Get the order of the categorical horizons for plotting
-
plot_benchmarks()
- Plot benchmark summaries
-
plot_components()
- Get plot components (colors for now)
-
plot_crps_underlay()
- Make CRPS underlay figure
-
plot_forecast_comparison_t()
- Compare two hospital admissions forecasts over time.
-
plot_heatmap_relative_wis()
- Make a heatmap of relative WIS across locations
-
plot_hub_performance_by_period()
- Get plot of overall hub performance, grouped by period
-
plot_model_hosp_t_comparison()
- Make a plot comparing the fit and forecasted hospital admissions from the wastewater-informed and hospital admissions-only models.
-
plot_pred_actual_hosp()
- Make a plot comparing the fit and forecasted hospital admissions from the wastewater and hospital admissions model for a few example states
-
plot_pred_actual_ww()
- Make concentration fit and forecast figure
-
plot_rel_crps_by_horizon()
- Make head to head CRPS distribution comparison plot for a single location
-
plot_rel_score_distribution()
- Get a dotsinterval plot of a relative score distribution for individual location/forecast-date forecast problems.
-
plot_rel_score_dists()
- Make a dotsinterval plot of relative scores stratified by specified variables.
-
plot_rel_score_dists_by_horizon()
- Make figure that stratifies across location and forecast dates
-
plot_rel_score_heatmap()
- Plot a heatmap of the relative crps by locations and forecast date for the head-to-head comparison
-
plot_score_decomposed_bars()
- Barplot of WIS or CRPS decomposed into overprediction, underprediction, and dispersion
-
plot_score_scatter()
- Plot a scatterplot of scores comparing those of one model to those of another, for a given stratification.
-
plot_score_t()
- Plotting functions for absolute forecast scores Plot average score over time for model comparison
-
plot_scores_by_scenario()
- Barplot of scores by scenario
-
plot_total_admissions()
- Make a figure of overall admissions (summed across locations) for context
-
plot_ww_conc_by_site()
- Make concentration fit and forecast figure
-
postprocess_successful_fit()
- Postprocess a successful eval fitting job.
-
process_recent_trend_fits()
- Process BRMS log-linear fits of recent trends
-
pull_and_write_hosp_data()
- Pull and save the current state of NHSN data on HealthData.gov
-
pull_daily_nhsn()
- Pull daily NHSN data from HealthData.gov
-
pull_hub_forecasts()
- Pull hub submissions and create a scorable table as the output of
scoringutils::as_forecast_quantile()
.
-
quantile_metrics
- Metrics to use for quantile scores
-
query_and_select_models()
- Query Zoltar for models to include in the analysis
-
read_rds_with_suffix()
- Load an object from an
.rds
serialized file in a given output_dir
, with programmatic naming.
-
relative_wis_histogram()
- Plot a histogram of individual forecast relative WIS values
-
sample_metrics
- Metrics to use for quantile scores
-
save_rds_with_suffix()
- Save an object to .rds in a given
dir
, with programmatic naming.
-
save_table()
- Save table
-
scale_alpha_score_component()
- ggplot scale for fill alphas by name of WIS/CRPS score component.
-
scale_fill_model()
scale_color_model()
- ggplot discrete scales for forecast models
-
score_hub_forecasts()
- Score hub forecasts
-
score_quantiles()
- Get the scores for every day for a location, forecast date, and scenario from the quantiles during the forecast period
-
score_real_time_outputs()
- Load in and score the real-time outputs
-
score_samples()
- Get the scores for every day for a particular location and forecast date
-
select_like()
- Select columns from one dataframe based on the column spec of a second dataframe
-
soql_is_in()
- Return a
soql::soql_where()
construct for a given column being in a list of values
-
summarize_std_rank()
- Summarize standardize rank with medians and 25th,75th percentiles
-
three_location_forecast_fig()
- Make a figure showing three example hospital admissions forecast timeseries and associated wastewater concentration timeseries.
-
with_dependencies()
- Helper function to note targets dependencies not explicitly noted in the function call