save_model {waves}

Save spectral prediction model and model performance statistics

Description

Given a set of pretreatment methods, saves the best spectral prediction model and model statistics to model.save.folder as model.name.Rds and model.name_stats.csv respectively. If only one pretreatment method is supplied, results from that method are stored.

Usage

save_model(
  df,
  write.model = TRUE,
  pretreatment = 1,
  model.save.folder = NULL,
  model.name = "PredictionModel",
  best.model.metric = "RMSE",
  k.folds = 5,
  proportion.train = 0.7,
  tune.length = 50,
  model.method = "pls",
  num.iterations = 10,
  stratified.sampling = TRUE,
  cv.scheme = NULL,
  trial1 = NULL,
  trial2 = NULL,
  trial3 = NULL,
  seed = 1,
  verbose = TRUE,
  save.model = deprecated(),
  wavelengths = deprecated(),
  autoselect.preprocessing = deprecated(),
  preprocessing.method = deprecated()
)

Arguments

df

data.frame object. First column contains unique identifiers, second contains reference values, followed by spectral columns. Include no other columns to right of spectra! Column names of spectra must start with "X" and reference column must be named "reference"

write.model

If TRUE, the trained model will be saved in .Rds format to the location specified by model.save.folder. If FALSE, the best model will be output by the function but will not save to a file. Default is TRUE.

pretreatment

Number or list of numbers 1:13 corresponding to desired pretreatment method(s):

1. Raw data (default)

2. Standard normal variate (SNV)

3. SNV and first derivative

4. SNV and second derivative

5. First derivative

6. Second derivative

7. Savitzky–Golay filter (SG)

8. SNV and SG

9. Gap-segment derivative (window size = 11)

10. SG and first derivative (window size = 5)

11. SG and first derivative (window size = 11)

12. SG and second derivative (window size = 5)

13. SG and second derivative (window size = 11)

model.save.folder

Path to folder where model will be saved. If not provided, will save to working directory.

model.name

Name that model will be saved as in model.save.folder. Default is "PredictionModel".

best.model.metric

Metric used to decide which model is best. Must be either "RMSE" or "Rsquared"

k.folds

Number indicating the number of folds for k-fold cross-validation during model training. Default is 5.

proportion.train

Fraction of samples to include in the training set. Default is 0.7.

tune.length

Number delineating search space for tuning of the PLSR hyperparameter ncomp. Must be set to 5 when using the random forest algorithm (model.method == rf). Default is 50.

model.method

Model type to use for training. Valid options include:

• "pls": Partial least squares regression (Default)

• "rf": Random forest

• "svmLinear": Support vector machine with linear kernel

• "svmRadial": Support vector machine with radial kernel

num.iterations

Number of training iterations to perform

stratified.sampling

If TRUE, training and test sets will be selected using stratified random sampling. This term is only used if test.data == NULL. Default is TRUE.

cv.scheme

A cross validation (CV) scheme from Jarquín et al., 2017. Options for cv.scheme include:

• "CV1": untested lines in tested environments

• "CV2": tested lines in tested environments

• "CV0": tested lines in untested environments

• "CV00": untested lines in untested environments

trial1

data.frame object that is for use only when cv.scheme is provided. Contains the trial to be tested in subsequent model training functions. The first column contains unique identifiers, second contains genotypes, third contains reference values, followed by spectral columns. Include no other columns to right of spectra! Column names of spectra must start with "X", reference column must be named "reference", and genotype column must be named "genotype".

trial2

data.frame object that is for use only when cv.scheme is provided. This data.frame contains a trial that has overlapping genotypes with trial1 but that were grown in a different site/year (different environment). Formatting must be consistent with trial1.

trial3

data.frame object that is for use only when cv.scheme is provided. This data.frame contains a trial that may or may not contain genotypes that overlap with trial1. Formatting must be consistent with trial1.

seed

Integer to be used internally as input for set.seed(). Only used if stratified.sampling = TRUE. In all other cases, seed is set to the current iteration number. Default is 1.

verbose

If TRUE, the number of rows removed through filtering will be printed to the console. Default is TRUE.

save.model

DEPRECATED save.model = FALSE is no longer supported; this function will always return a saved model.

wavelengths

DEPRECATED wavelengths is no longer supported; this information is now inferred from df column names

autoselect.preprocessing

DEPRECATED autoselect.preprocessing = FALSE is no longer supported. If multiple pretreatment methods are supplied, the best will be automatically selected as the model to be saved.

preprocessing.method

DEPRECATED preprocessing.method has been renamed "pretreatment"

Details

Wrapper that uses pretreat_spectra, format_cv, and train_spectra functions.

Value

List of model stats (in data.frame) and trained model object. If the parameter write.model is TRUE, both objects are saved to model.save.folder. To use the optimally trained model for predictions, use tuned parameters from $bestTune.

Author(s)

Jenna Hershberger [e-mail]

Examples

library(magrittr)
test.model <- ikeogu.2017 %>%
  dplyr::filter(study.name == "C16Mcal") %>%
  dplyr::rename(reference = DMC.oven,
                unique.id = sample.id) %>%
  dplyr::select(unique.id, reference, dplyr::starts_with("X")) %>%
  na.omit() %>%
  save_model(
    df = .,
    write.model = FALSE,
    pretreatment = 1:13,
    model.name = "my_prediction_model",
    tune.length = 3,
    num.iterations = 3
  )
summary(test.model$best.model)
test.model$best.model.stats