Man pages for vidarsumo/sumots
Helper functions for Sumo Analytics

accuracy_by_idFind best model per id and create forecast out of sample
best_weighted_averageFind best weights for each model in an ensemble
create_covidCreates dummy variable for Covid
create_holidayCreates holiday variables to be used in modeling
data_prep_funcPrepares data for modeling
data_prep_single_tsA function to prepare a single time series for forecasting...
finalize_deeparFunction to select best DeepAR based on chosen metric
finalize_nbeatsFunction to select best N-BEATS based on chosen metric
get_gmm_clustersFunction to create clusters using time series features and...
isl_holidaysHolidays and other events in Iceland
ml_models_wflwA function to train simple univariate algorithms
ml_tuneFunction to tune ML algos for multiple time series...
prepare_energy_dataPrepares data for energy forecasting
stack_ml_modelsCreates super-learner by stacking differnt ML models
sumo_forecast_per_idAssign a different model to each time series
train_simple_modelsA function to train simple univariate algorithms
tune_deeparFunction to tune DeepAR
tune_nbeatsFunction to tune N-BEATS
wflw_creatorA helper function to create workflow passed to the ml_tune()...
vidarsumo/sumots documentation built on June 29, 2021, 4:23 a.m.