Package index • beethoven

Utility

Cleaning and adjusting data structure for succinct pipeline

attach_pred(): Combine prediction values from base learners

attach_xy(): Attach XY coordinates to a data frame

feature_raw_download(): Check file status and download if necessary

fl_dates(): Extract the first and last elements of a list

load_modis_files(): Load MODIS files from a specified path.

loadargs(): Load arguments from the formatted argument list file

pred_colname(): Assign column name to base learner prediction based on hyperparameters.

read_locs(): Read AQS data

read_paths(): Read paths from a directory with a specific file extension

reduce_list(): Combine dynamically branched sublists based on common column names

set_args_calc(): Set arguments for the calculation process

set_args_download(): Generate argument list for raw data download

set_slurm_resource(): Set resource management for SLURM

set_target_years(): Set which years to be processed

split_dates(): Split a date range into subranges

unmarshal_function(): Unmarshal functions

Calculation

Main calculation functions for branching, parallel processing, and parameter injection

calc_geos_strict(): Process atmospheric composition data by chunks

calc_gmted_direct(): Reflown gmted processing

calc_narr2(): Calculate aggregated values for specified locations

calculate(): Spatiotemporal covariate calculation

inject_calculate(): Injects the calculate function with specified arguments.

inject_geos(): Injects geographic information into a data frame

inject_gmted(): Injects GMTED data into specified locations

inject_match(): Injects the calculate function with matched arguments.

inject_modis_par(): Injects arguments to parallelize MODIS/VIIRS data processing

inject_nlcd(): Inject arguments into NLCD calculation function for branching

par_narr(): Parallelize NARR feature calculation

process_counties(): Load county sf object

process_geos_bulk(): Process atmospheric composition data by chunks (v2)

process_narr2(): Process NARR Data (v2)

query_modis_files(): Identify MODIS files

Post-calculation

Merging calculation results

add_time_col(): Add Time Column

append_predecessors(): Append Predecessors

impute_all(): Impute missing values and attach lagged features

post_calc_autojoin(): Automatic joining by the time and spatial identifiers

post_calc_convert_time(): Convert time column to character

post_calc_df_year_expand(): Expand a data frame by year

post_calc_drop_cols(): Remove columns from a data frame based on regular expression patterns.

post_calc_join_yeardate(): Join a data.frame with a year-only date column to that with a full date column

post_calc_merge_all(): Merge spatial and spatiotemporal covariate data

post_calc_merge_features(): Merge input data.frame objects

post_calc_unify_timecols(): Change time column name

post_calc_year_expand(): Map the available raw data years over the given period

reduce_merge(): Reduce and merge a list of data tables

Base learner

Fitting base learner and helper functions

assign_learner_cv(): Shuffle cross-validation mode for each learner type

convert_cv_index_rset(): Generate manual rset object from spatiotemporal cross-validation indices

fit_base_learner(): Base learner: tune hyperparameters and retrieve the best model

fit_base_tune(): Tune base learner

generate_cv_index_sp(): Prepare spatial and spatiotemporal cross validation sets

generate_cv_index_spt(): Generate spatio-temporal cross-validation index with anticlust

generate_cv_index_ts(): Generate temporal cross-validation index

make_subdata(): Make sampled subdataframes for base learners

switch_generate_cv_rset(): Choose cross-validation strategy for the base learner

switch_model(): Define a base learner model based on parsnip and tune

vis_spt_rset(): Visualize the spatio-temporal cross-validation index

Meta learner

Fitting meta learner

fit_meta_learner(): Fit meta learner

predict_meta_learner(): Predict meta learner

Miscellaneous

Miscellaneous functions for general use

divisor(): Get Divisors