Package index
Data
Example datasets included with the package, used to demonstrate and test deweathering functions.
-
aqroadside - Example air quality monitoring data for openair
Tune
Tune hyperparameters for a deweathering model before it is fit. A ‘best’ parameter set is automatically determined, but other functions are provided to allow for closer interrogation so that these can be refined.
-
tune_dw_model() - Tune a deweather model
-
plot_tdw_tuning_metrics() - Plot Tuning Metrics from
tune_dw_model() -
plot_tdw_testing_scatter() - Plot Observed vs Modelled Scatter using the 'best parameters' from
tune_dw_model()
Build
Core functions for fitting deweathering models, used throughout the rest of the deweather package for interpretation and prediction.
-
build_dw_model() - Build a Deweather Model
-
finalise_tdw_model() - Use the 'best parameters' determined by
tune_dw_model()to build a Deweather Model -
append_dw_vars() - Conveniently append common 'deweathering' variables to an air quality time series
Examine
‘getters’ to extract specific features of a built deweather model or a deweather ‘tuning’ object.
-
get_dw_pollutant()get_dw_vars()get_dw_engine()get_dw_params()get_dw_input_data()get_dw_model()get_dw_importance() - Getters for various deweather model features
-
get_tdw_pollutant()get_tdw_vars()get_tdw_engine()get_tdw_best_params()get_tdw_tuning_metrics()get_tdw_testing_metrics()get_tdw_testing_data() - Getters for various deweather tuning object features
Visualise
Functions for visualizing model components and relationships, including variable importance and partial dependence plots.
-
plot_dw_importance() - Visualise deweather model feature importance
-
plot_dw_partial_1d() - Create partial dependence plots for deweather models
-
plot_dw_partial_2d() - Create a 2-way partial dependence plot for deweather models
-
predict_dw() - Use a deweather model to predict with a new dataset
-
simulate_dw_met() - Function to run random meteorological simulations on a deweather model