Package index • waves

Core model development workflow

These functions prepare spectra for model development and allow for the streamlined testing of model performance.

plot_spectra(): Plot spectral data, highlighting outliers as identified using Mahalanobis distance
filter_spectra(): Filter spectral data frame based on Mahalanobis distance
aggregate_spectra(): Aggregate data based on grouping variables and a user-provided function
test_spectra(): Test the performance of spectral models

Internal use

While these functions can be used on their own, they were developed to be called within test_spectra().

pretreat_spectra(): Pretreat spectral data according to user-designated method
format_cv(): Format multiple trials with or without overlapping genotypes into training and test sets according to user-provided cross validation scheme
train_spectra(): Train a model based predict reference values with spectral data

Model development and testing for routine use

save_model(): Save spectral prediction model and model performance statistics
predict_spectra(): Use provided model object to predict trait values with input dataset

Example Dataset

ikeogu.2017: Example vis-NIRS and reference dataset