Function reference • cfaforecastrenewalww

All functions

add_pmfs(): Add probability mass functions

aggregate_ww(): Aggregate the WW data to the state-level

boxplot_whiskers(): Get the upper and lower ends of the whiskers of a single boxplot of data in x

categorical_entropy(): Entropy of sample from categorical distribution

cleanup_secrets(): Cleans up environment when secrets and API information is no longer needed

compile_model(): Compile a stan model while pointing at the package default include directory (stan) for #include statements

convert_to_logmean(): Convert to logmean in lognorm distribution

convert_to_logsd(): Convert to logsd in lognormal distribution

create_dir(): Create a new directory if one doesn't exist

drop_first_and_renormalize(): Drop the first element of a simplex

entropy(): Entropy of probability mass function p

example_df: Input dataframe for fitting the site-level infection dynamics wastewater model

example_params: Example model parameters and hyperparameters for priors

fit_aggregated_model(): Fit aggregated model

fit_model(): Fit site level model

fit_site_level_model(): Fit site-level model

flag_ww_outliers(): Flag WW outliers

format_hosp_draws_hub_sub(): Prepare posterior draws of hospital admissions for the COVID-19 forecast hub

generate_simulated_data(): Generate simulated data from the underlying model's generative process

get_all_quantiles(): Get all quantiles

get_all_training_data(): Get long tibble of data needed for model fitting

get_combo_quantile_plot(): Title

get_config_vals(): Read in the config yaml and add run-specific dynamic configuration to it.

get_df_for_hub_submission(): Load in future hosp draws and combine into a single dataframe

get_df_of_filepaths(): This code was adapted from code written (under an MIT license) as part of the epinowcast package (https://github.com/epinowcast/epinowcast) Get dataframe of filepaths

get_diagnostics(): Get a data frame of data and model fit diagnostics

get_figure_file_path(): Get the figure file path

get_figure_output_subdirectory(): Get the path to the subdirectory in which to save figures for a given top-level output directory

get_full_param_distrib(): Get full parameter distribution

get_gen_quants_draws(): Get generated quantities draws site level model

get_generated_quantities_draws(): Get generated quantities draws

get_generation_interval_gamma(): Get the generation interval from a gamma distribution

get_generation_interval_lnorm(): Get the generation interval from a lognormal distribution

get_hosp_data(): Gets hospitalization data that model will be trained on

get_hosp_reporting_delay(): Get the hospital reporting delay

get_ind_m(): Get index matrix

get_ind_m_cum_sum(): Get index matrix

get_init_fun(): Get initialization function for stan

get_loc_model_map(): Get location to model map

get_location_notes_table(): Get a table that records notes on data characteristics and model choices for individual locations, e.g. the absence of recent wastewater observations or the decision to publish the hospital admissions only model

get_low_case_count_diagnostic(): Calculate low case count diagnostic flag

get_metadata(): Get metadata from a single group training data

get_metadata_yaml(): Create YAML formatted metadata file listing states without wastewater data, states with insufficient wastewater data, and states for which we used the hospital admissions only model.

get_model_file_name(): Get shortened name for writing a file with model file name

get_model_file_path(): Get path to stan wastewater model

get_model_param_df(): Information about a ww model's parameters and generated quantities

get_model_run_diagnostics(): Get the parameters that were flagged for diagnostics for the model run

get_nearest_forecast_date(): Local functions that are used to pre-process our specific data for use in the stan model These will be used to get the data up to the point it needs to be in the get_stan_data functions, which format the data into the stan lists that the stan model needs. None of these will be exported Get forecast date from today's date

get_param_samples_long_df(): Get full parameter distribution

get_parameter_draws(): Get posterior parameter draws and summary stats

get_params(): Get parameters for model run

get_pars(): Get parameters

get_pdf_file_path(): get the cleaned pdf filepath

get_pdf_output_subdirectory(): Get the path to the subdirectory in which to save summary output for a given top-level output directory.

get_pipeline_metadata(): Assemble pipeline metadata for a given run, returning it as a list and writing it to disk as a YAML file.

get_plot_change_in_p_hosp_t(): Get plot of the change in IHR over time

get_plot_covidhub_submission(): Get plot COVID hub submission

get_plot_draws(): Plot of draws for a single forecast date, with optional groupings

get_plot_labsite_ww_draws(): Plot of lab-site level wastewater draws

get_plot_metadata(): Get the metadata needed for plotting

get_plot_p_hosp_t(): Get plot of IHR(t)

get_plot_param_distribs(): Get plot of parameter distributions

get_plot_single_param(): Get plot single parameter

get_ppa_long(): Reformats output of posterior_predictive_analysis() into something more easily usable

get_prior_pred_hosps_df(): Observables extraction from prior predictive samples

get_prior_pred_ww_df(): Observables extraction from prior predictive samples

get_quantiles(): Get the quantiles being used

get_quantiles_on_site_level_ww(): Quantile the site level expected observed WW

get_random_string(): Get a random string of length 8 containing 3 letters, 4 digits, 1 letter

get_raw_param_draws(): Get raw parameter draws

get_regions_for_mapping(): : get states mapped to their regions

get_relative_forecast_dir(): Hardcoded figure filepath for figures per the read me

get_repo_forecast_file_path(): Get the filepath to save the forecasts within this repo

get_rt_boxplot_across_states(): Get R(t) box plot across states

get_rt_from_draws(): Get R(t) for a single location

get_rt_site_level(): Plot of site-level Rt

get_rt_subpop_level(): Plot of site-level Rt

get_scores(): Get scores

get_secret(): Interface for getting information needed for API interfacing

get_site_county_map(): Get site county mapp

get_stan_data(): Get stan data

get_stan_data_site_level_model(): Get stan data site level model

get_state_level_hosp_data(): Get raw hospitalization data

get_state_level_summary(): Get summary of WW viral concentrations by state

get_submission_file_path(): Get the hardcoded submission file path

get_submission_filepath_df(): Get a consolidated dataframe with each location and its corresponding file path for the necesary outputs.

get_summarized_table(): Get summarized table with each location and

get_summary_stats(): Get sumamry stats of diagnostics

get_testing_data(): Get testing dataset

get_training_data(): Get dataframe with wastewater and hospitalizaiton data for training the model

get_wastewater_diagnostic(): Get some basic flags on the wastewater data, for internal use and for reporting out in our metadata

get_weekly_summary(): Summarize data into weekly by site

get_ww_data(): Get the watstewater data for a particular source and geographic granularity

get_ww_site_plots(): Plot of lab-site level wastewater quantiles

init_subset_nwss_data(): Initial subsetting of NWSS data

loc_abbr_to_flusight_code(): Convert a two-letter location abbreviation to a two-character FluSight location code

log_standardize(): As standardize, but on log scale, possibly removing zeros (or near-zeros)

make_hospital_onset_delay_pmf(): Make hospital onset delay pmf

make_incubation_period_pmf(): Make incubation period pmf

make_pps_data(): Gets generated quantites from a fitted ww model and formats for stan

make_reporting_delay_pmf(): Make reporting delay pmf

manual_removal_of_hosp_data(): Manual removal of hopsital admissions data points in particular states

max_delta(): Maximum finite difference of x

nhsn_soda_query(): Construct a Socrata open data API (SODA) query for the NSHN dataset

param_df: Input parameters used to generate example_df

pivot_data_for_plotting(): Pivot the quantiles output for easy plotting

plot_combined_data(): Plot the WW and hospitalization data together

plot_quantiles(): Plot a timeseries of quantiles

posterior_predictive_analysis(): Runs posterior- and prior--predictive analyses and returns summaries of performance

prior_predictive_check(): Draws data under the model's prior.

pull_nhsn(): Pull NHSN data from HealthData.gov

read_diagnostics_df(): Load in and combine all the diagnostic tables

render_diagnostic_report(): Render diagnostic report

save_timestamped_nwss_data(): Save time stampped NWSS data

save_to_pdf(): Save a single pdf of plots

setup_secrets(): Handles values needed to retrieve data from APIs to run pipeline

simulate_double_censored_pmf(): Simulate daily double censored PMF. From epinowcast: https://package.epinowcast.org/dev/reference/simulate_double_censored_pmf.html

site_level_inf_inits(): Get initialization function for stan for site-level infection model

soql_is_in(): Return a soql::soql_where() construct for a given column being in a list of values

standardize(): x -> (x - location(x))/dispersion(x) location is mean, dispersion is SD or MAD

state_agg_inits(): Get initialization function for stan for aggregated model

subsample_sites(): Subsample wastewater sites

summarize_scores(): Summarize scores

trajectories_to_quantiles(): Aggregate individual trajectory timeseries or forecasts to quantile timeseries or forecasts

validate_paramlist(): Validate a parameter list

varies(): Is there any variation in x?

ww_model(): This code was adapted from code written (under an MIT license) as part of the `epinowcast` package (https://github.com/epinowcast/epinowcast) ww_model

zoom_boxplot_y(): Get y range required to zoom a boxplot in to ignore outliers df is the data, x the grouping variable, y the variable to be boxplotted