Help for package waves

Title:

Vis-NIR Spectral Analysis Wrapper

Version:

0.2.6

Maintainer:

Jenna Hershberger <jmh579@cornell.edu>

Description:

Originally designed application in the context of resource-limited plant research and breeding programs, 'waves' provides an open-source solution to spectral data processing and model development by bringing useful packages together into a streamlined pipeline. This package is wrapper for functions related to the analysis of point visible and near-infrared reflectance measurements. It includes visualization, filtering, aggregation, preprocessing, cross-validation set formation, model training, and prediction functions to enable open-source association of spectral and reference data. This package is documented in a peer-reviewed manuscript in the Plant Phenome Journal <doi:10.1002/ppj2.20012>. Specialized cross-validation schemes are described in detail in Jarquín et al. (2017) <doi:10.3835/plantgenome2016.12.0130>. Example data is from Ikeogu et al. (2017) <doi:10.1371/journal.pone.0188918>.

License:

MIT + file LICENSE

URL:

https://GoreLab.github.io/waves/, https://github.com/GoreLab/waves

BugReports:

https://github.com/GoreLab/waves/issues

Depends:

R (≥ 3.5)

Imports:

caret, dplyr, ggplot2, lifecycle, magrittr, pls, prospectr, randomForest, readr, rlang, scales, spectacles, stringr, tibble, tidyr (≥ 1.0), tidyselect

Suggests:

testthat (≥ 2.1.0), knitr, rmarkdown

Encoding:

UTF-8

LazyData:

true

RoxygenNote:

7.3.2

VignetteBuilder:

knitr, rmarkdown

NeedsCompilation:

Packaged:

2025-10-30 22:46:12 UTC; Jenna

Author:

Jenna Hershberger

[aut, cre], Michael Gore [ths], NSF BREAD IOS-1543958 [fnd]

Repository:

CRAN

Date/Publication:

2025-10-30 23:50:02 UTC

Aggregate data based on grouping variables and a user-provided function

Description

Use grouping variables to collapse spectral data.frame by mean or median. Recommended for use after filter_spectra

Usage

aggregate_spectra(df, grouping.colnames, reference.value.colname,
  agg.function)

Arguments

df

data.frame object containing one or multiple columns of grouping variables (must be consistent within each group), column of reference values (optional), and columns of spectra. Spectral column names must start with "X".

grouping.colnames

Names of columns to be used as grouping variables. Minimum 2 variables required. Default is c("trial", "plot").

reference.value.colname

Name of reference column to be aggregated along with spectra. Default is "reference"

agg.function

Name of function (string format) to be used for sample aggregation. Must be either "mean" or "median". Default is "mean".

Value

data.frame object df aggregated based on grouping column by agg.function

Author(s)

Jenna Hershberger jmh579@cornell.edu

Examples

library(magrittr)
aggregated.test <- ikeogu.2017 %>%
  dplyr::select(-TCC) %>%
  na.omit() %>%
  aggregate_spectra(
    grouping.colnames = c("study.name"),
    reference.value.colname = "DMC.oven",
    agg.function = "mean"
  )
aggregated.test[1:5, 1:5]

Filter spectral data frame based on Mahalanobis distance

Description

Determine Mahalanobis distances of observations (rows) within a given data.frame with spectral data. Option to filter out observations based on these distances.

Usage

filter_spectra(df, filter, return.distances, num.col.before.spectra,
  window.size, verbose)

Arguments

df

data.frame object containing columns of spectra and rows of observations. Spectral columns must be labeled with an "X" and then the wavelength (example: "X740" = 740nm). Left-most column must be unique ID. May also contain columns of metadata between the unique ID and spectral columns. Cannot contain any missing values. Metadata column names may not start with "X".

filter

boolean that determines whether or not the input data.frame will be filtered. If TRUE, df will be filtered according to squared Mahalanobis distance with a 95% cutoff from a chi-square distribution with degrees of freedom = number of spectral columns. If FALSE, a column of squared Mahalanobis distances h.distance will be added to the right side of df and all rows will be returned. Default is TRUE.

return.distances

boolean that determines whether a column of squared Mahalanobis distances will be included in output data.frame. If TRUE, a column of Mahalanobis distances for each row will be added to the right side of df. Default is FALSE.

num.col.before.spectra

number of columns to the left of the spectral matrix in df. Default is 4.

window.size

number defining the size of window to use when calculating the covariance of the spectra (required to calculate Mahalanobis distance). Default is 10.

verbose

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

Details

This function uses a chi-square distribution with 95% cutoff where degrees of freedom = number of wavelengths (columns) in the input data.frame.

Value

If filter is TRUE, returns filtered data frame df and reports the number of rows removed. The Mahalanobis distance with a cutoff of 95% of chi-square distribution (degrees of freedom = number of wavelengths) is used as filtering criteria. If filter is FALSE, returns full input df with column h.distances containing the Mahalanobis distance for each row.

Author(s)

Jenna Hershberger jmh579@cornell.edu

References

Johnson, R.A., and D.W. Wichern. 2007. Applied Multivariate Statistical Analysis (6th Edition). pg 189

Examples

library(magrittr)
filtered.test <- ikeogu.2017 %>%
  dplyr::select(-TCC) %>%
  na.omit() %>%
  filter_spectra(
    df = .,
    filter = TRUE,
    return.distances = TRUE,
    num.col.before.spectra = 5,
    window.size = 15
  )
filtered.test[1:5, c(1:5, (ncol(filtered.test) - 5):ncol(filtered.test))]

Format multiple trials with or without overlapping genotypes into training and test sets according to user-provided cross validation scheme

Description

Standalone function that is also used within train_spectra to divide trials or studies into training and test sets based on overlap in trial environments and genotype entries

Usage

format_cv(
  trial1,
  trial2,
  trial3 = NULL,
  cv.scheme,
  stratified.sampling = TRUE,
  proportion.train = 0.7,
  seed = NULL,
  remove.genotype = FALSE
)

Arguments

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.

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

stratified.sampling

If TRUE, training and test sets will be selected using stratified random sampling. Default is TRUE.

proportion.train

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

seed

Number used in the function set.seed() for reproducible randomization. If NULL, no seed is set. Default is NULL.

remove.genotype

boolean that, if TRUE, removes the "genotype" column is removed from the output data.frame. Default is FALSE.

Details

Use of a cross-validation scheme requires a column in the input data.frame named "genotype" to ensure proper sorting of training and test sets. Variables trial1 and trial2 are required, while trial 3 is optional.

Value

List of data.frames ($train.set, $test.set) compiled according to user-provided cross validation scheme.

Author(s)

Jenna Hershberger jmh579@cornell.edu

References

Jarquín, D., C. Lemes da Silva, R. C. Gaynor, J. Poland, A. Fritz, R. Howard, S. Battenfield, and J. Crossa. 2017. Increasing genomic-enabled prediction accuracy by modeling genotype × environment interactions in Kansas wheat. Plant Genome 10(2):1-15. <doi:10.3835/plantgenome2016.12.0130>

Examples

# Must have a column called "genotype", so we'll create a fake one for now
# We will use CV00, which does not require any overlap in genotypes
# In real scenarios, CV schemes that rely on genotypes should not be applied
# when genotypes are unknown, as in this case.
library(magrittr)
trials <- ikeogu.2017 %>%
  dplyr::mutate(genotype = 1:nrow(ikeogu.2017)) %>% # fake for this example
  dplyr::rename(reference = DMC.oven) %>%
  dplyr::select(
    study.name, sample.id, genotype, reference,
    tidyselect::starts_with("X")
  )
trial1 <- trials %>%
  dplyr::filter(study.name == "C16Mcal") %>%
  dplyr::select(-study.name)
trial2 <- trials %>%
  dplyr::filter(study.name == "C16Mval") %>%
  dplyr::select(-study.name)
cv.list <- format_cv(
  trial1 = trial1, trial2 = trial2, cv.scheme = "CV00",
  stratified.sampling = FALSE, remove.genotype = TRUE
)
cv.list$train.set[1:5, 1:5]
cv.list$test.set[1:5, 1:5]

Example vis-NIRS and reference dataset

Description

The 'ikeogu.2017' data set contains raw vis-NIRS scans, total carotenoid content, and cassava root dry matter content (using the oven method) from the 2017 PLOS One paper by Ikeogu et al. This dataset contains a subset of the original scans and reference values from the supplementary files of the paper. 'ikeogu.2017' is a 'data.frame' that contains the following columns:

study.name = Name of the study as described in Ikeogu et al. (2017).
sample.id = Unique identifier for each individual root sample
DMC.oven = Cassava root dry matter content, the percentage of dry weight relative to fresh weight of a sample after oven drying.
TCC = Total carotenoid content (\mu g/g, unknown whether on a fresh or dry weight basis) as measured by high performance liquid chromatography
X350:X2500 = spectral reflectance measured with the QualitySpec Trek: S-10016 vis-NIR spectrometer. Each cell represents the mean of 150 scans on a single root at a single wavelength.

Usage

ikeogu.2017

Format

An object of class tbl_df (inherits from tbl, data.frame) with 175 rows and 2155 columns.

Author(s)

Original authors: Ikeogu, U.N., F. Davrieux, D. Dufour, H. Ceballos, C.N. Egesi, and J. Jannink. Reformatted by Jenna Hershberger.

References

Ikeogu, U.N., F. Davrieux, D. Dufour, H. Ceballos, C.N. Egesi, et al. 2017. Rapid analyses of dry matter content and carotenoids in fresh cassava roots using a portable visible and near infrared spectrometer (Vis/NIRS). PLOS One 12(12): 1–17. doi: 10.1371/journal.pone.0188918.

Examples

library(magrittr)
library(ggplot2)
data(ikeogu.2017)
ikeogu.2017[1:10, 1:10]
ikeogu.2017 %>%
  dplyr::select(-starts_with("X")) %>%
  dplyr::group_by(study.name) %>%
  tidyr::gather(trait, value, c(DMC.oven:TCC), na.rm = TRUE) %>%
  ggplot2::ggplot(aes(x = study.name, y = value, fill = study.name)) +
  facet_wrap(~trait, scales = "free_y", nrow = 2) +
  geom_boxplot()

Plot spectral data, highlighting outliers as identified using Mahalanobis distance

Description

Generates a ggplot object of given spectra, with wavelength on the x axis and given spectral values on the y. Mahalanobis distance is used to calculate outliers, which are both identified on the plot. Rows from the original dataframe are printed to the console for each outlier that is identified.

Usage

plot_spectra(
  df,
  num.col.before.spectra = 1,
  window.size = 10,
  detect.outliers = TRUE,
  color = NULL,
  alternate.title = "",
  verbose = TRUE,
  wavelengths = deprecated()
)

Arguments

df

data.frame object containing columns of spectra. Spectral columns must be labeled with an "X" and then the wavelength (example: "X740" = 740nm). Left-most column must be unique ID. May also contain columns of metadata between the unique ID and spectral columns. Cannot contain any missing values. Metadata column names may not start with "X".

num.col.before.spectra

Number of columns to the left of the spectral matrix (including unique ID). Default is 1.

window.size

number defining the size of window to use when calculating the covariance of the spectra (required to calculate Mahalanobis distance). Default is 10.

detect.outliers

Boolean indicating whether spectra should be filtered before plotting. If TRUE, outliers are indicated by color in the resulting plot. If verbose is also set to TRUE, outlier metadata will be printed to the console. Default is TRUE.

color

String or vector of strings indicating colors to be passed to ggplot. Default is default ggplot colors.

alternate.title

String to be used as plot title. If detect.outliers is TRUE, a descriptive title will be supplied. If detect.outliers is FALSE, default is no title will be used.

verbose

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

wavelengths

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

Value

If verbose, prints unique ID and metadata for rows identified as outliers. Returns plot of spectral data with non-outliers in blue and outliers in red. X-axis is wavelengths and y-axis is spectral values.

Author(s)

Jenna Hershberger jmh579@cornell.edu

Examples


library(magrittr)
ikeogu.2017 %>%
  dplyr::rename(unique.id = sample.id) %>%
  dplyr::select(unique.id, dplyr::everything(), -TCC) %>%
  na.omit() %>%
  plot_spectra(
    df = .,
    num.col.before.spectra = 5,
    window.size = 15,
    detect.outliers = TRUE,
    color = NULL,
    alternate.title = NULL,
    verbose = TRUE
  )

Use provided model object to predict trait values with input dataset

Description

Loads an existing model and cross-validation performance statistics (created with save_model) and makes predictions based on new spectra.

Usage

predict_spectra(
  input.data,
  model.stats.location,
  model.location,
  model.method = "pls",
  wavelengths = deprecated()
)

Arguments

input.data

data.frame object of spectral data for input into a spectral prediction model. First column contains unique identifiers followed by spectral columns. Include no other columns to right of spectra! Column names of spectra must start with "X".

model.stats.location

String containing file path (including file name) to save location of "(model.name)_stats.csv" as output from the save_model function.

model.location

String containing file path (including file name) to location where the trained model ("(model.name).Rds") was saved as output by the save_model function.

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

wavelengths

DEPRECATED wavelengths is no longer supported; this information is now inferred from input.data column names

Value

data.frame object of predictions for each sample (row). First column is unique identifier supplied by input.data and second is predicted values

Author(s)

Jenna Hershberger jmh579@cornell.edu

Examples

## Not run: 
ikeogu.2017 %>%
  dplyr::select(sample.id, dplyr::starts_with("X")) %>%
  predict_spectra(
    input.data = .,
    model.stats.location = paste0(
      getwd(),
      "/my_model_stats.csv"
    ),
    model.location = paste0(getwd(), "/my_model.Rds")
  )

## End(Not run)

Pretreat spectral data according to user-designated method

Description

Pretreatment, also known as preprocessing, is often used to increase the signal to noise ratio in vis-NIR datasets. The waves function pretreat_spectra applies common spectral pretreatment methods such as standard normal variate and the Savitzky-Golay filter.

Usage

pretreat_spectra(
  df,
  test.data = NULL,
  pretreatment = 1,
  preprocessing.method = deprecated(),
  wavelengths = deprecated()
)

Arguments

df

data.frame object containing spectral data. First column(s) (optional) include metadata (with or without reference value column) followed by spectral columns. Spectral column names must be formatted as "X" followed by wavelength Include no other columns to right of spectra! No missing values permitted.

test.data

data.frame object with same format as train.data. Will be appended to df during pretreatment so that the same transformations are applied to each row. Default is NULL.

pretreatment

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

Raw data (default)
Standard normal variate (SNV)
SNV and first derivative
SNV and second derivative
First derivative
Second derivative
Savitzky–Golay filter (SG)
SNV and SG
Gap-segment derivative (window size = 11)
SG and first derivative (window size = 5)
SG and first derivative (window size = 11)
SG and second derivative (window size = 5)
SG and second derivative (window size = 11)

preprocessing.method

DEPRECATED preprocessing.method has been renamed "pretreatment"

wavelengths

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

Value

Pretreated df' (or list of data.frames) with reference column intact

Author(s)

Jenna Hershberger jmh579@cornell.edu

Examples

pretreat_spectra(df = ikeogu.2017, pretreatment = 3)[1:5, 1:5]

Functions renamed in waves 0.2.0

Description

‘r lifecycle::badge(’deprecated')'

waves 0.2.0 renamed a number of functions to ensure that every function name adheres to the tidyverse style guide.

* 'AggregateSpectra()' -> 'aggregate_spectra()' * 'DoPreprocessing()' -> 'pretreat_spectra()' * 'FilterSpectra()' -> 'filter_spectra()' * 'FormatCV()' -> 'format_cv()' * 'PlotSpectra()' -> 'plot_spectra()' * 'PredictFromSavedModel()' -> 'predict_spectra()' * 'SaveModel()' -> 'save_model()' * 'TestModelPerformance()' -> 'test_spectra()' * 'TrainSpectralModel()' -> 'train_spectra()'

Usage

AggregateSpectra(
  df,
  grouping.colnames = c("unique.id"),
  reference.value.colname = "reference",
  agg.function = "mean"
)

DoPreprocessing(df, test.data = NULL, pretreatment = 1)

FilterSpectra(
  df,
  filter = TRUE,
  return.distances = FALSE,
  num.col.before.spectra = 4,
  window.size = 10,
  verbose = TRUE
)

FormatCV(
  trial1,
  trial2,
  trial3 = NULL,
  cv.scheme,
  stratified.sampling = TRUE,
  proportion.train = 0.7,
  seed = NULL,
  remove.genotype = FALSE
)

PlotSpectra(
  df,
  num.col.before.spectra = 1,
  window.size = 10,
  detect.outliers = TRUE,
  color = NULL,
  alternate.title = NULL,
  verbose = TRUE
)

PredictFromSavedModel(
  input.data,
  model.stats.location,
  model.location,
  model.method = "pls"
)

SaveModel(
  df,
  save.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,
  verbose = TRUE
)

TestModelPerformance(
  train.data,
  num.iterations,
  test.data = NULL,
  pretreatment = 1,
  k.folds = 5,
  proportion.train = 0.7,
  tune.length = 50,
  model.method = "pls",
  best.model.metric = "RMSE",
  stratified.sampling = TRUE,
  cv.scheme = NULL,
  trial1 = NULL,
  trial2 = NULL,
  trial3 = NULL,
  split.test = FALSE,
  verbose = TRUE
)

TrainSpectralModel(
  df,
  num.iterations,
  test.data = NULL,
  k.folds = 5,
  proportion.train = 0.7,
  tune.length = 50,
  model.method = "pls",
  best.model.metric = "RMSE",
  stratified.sampling = TRUE,
  cv.scheme = NULL,
  trial1 = NULL,
  trial2 = NULL,
  trial3 = NULL,
  split.test = FALSE,
  verbose = TRUE
)

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):

Raw data (default)
Standard normal variate (SNV)
SNV and first derivative
SNV and second derivative
First derivative
Second derivative
Savitzky–Golay filter (SG)
SNV and SG
Gap-segment derivative (window size = 11)
SG and first derivative (window size = 5)
SG and first derivative (window size = 11)
SG and second derivative (window size = 5)
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

trial2

trial3

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 jmh579@cornell.edu

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

Test the performance of spectral models

Description

Wrapper that trains models based spectral data to predict reference values and reports model performance statistics

Usage

test_spectra(
  train.data,
  num.iterations,
  test.data = NULL,
  pretreatment = 1,
  k.folds = 5,
  proportion.train = 0.7,
  tune.length = 50,
  model.method = "pls",
  best.model.metric = "RMSE",
  stratified.sampling = TRUE,
  cv.scheme = NULL,
  trial1 = NULL,
  trial2 = NULL,
  trial3 = NULL,
  split.test = FALSE,
  seed = 1,
  verbose = TRUE,
  wavelengths = deprecated(),
  preprocessing = deprecated(),
  output.summary = deprecated(),
  rf.variable.importance = deprecated()
)

Arguments

train.data

data.frame object of spectral data for input into a spectral prediction model. 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".

num.iterations

Number of training iterations to perform

test.data

data.frame with same specifications as df. Use if specific test set is desired for hyperparameter tuning. If NULL, function will automatically train with a stratified sample of 70%. Default is NULL.

pretreatment

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

Raw data (default)
Standard normal variate (SNV)
SNV and first derivative
SNV and second derivative
First derivative
Second derivative
Savitzky–Golay filter (SG)
SNV and SG
Gap-segment derivative (window size = 11)
SG and first derivative (window size = 5)
SG and first derivative (window size = 11)
SG and second derivative (window size = 5)
SG and second derivative (window size = 11)

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

best.model.metric

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

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

trial2

trial3

split.test

boolean that allows for a fixed training set and a split test set. Example// train model on data from two breeding programs and a stratified subset (70%) of a third and test on the remaining samples (30%) of the third. If FALSE, the entire provided test set test.data will remain as a testing set or if none is provided, 30% of the provided train.data will be used for testing. Default is FALSE.

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.

wavelengths

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

preprocessing

DEPRECATED please use pretreatment to specify the specific pretreatment(s) to test. For behavior identical to that of preprocessing = TRUE, set pretreatment = 1:13'.

output.summary

DEPRECATED output.summary = FALSE is no longer supported; a summary of output is always returned alongside the full performance statistics.

rf.variable.importance

DEPRECATED rf.variable.importance = FALSE is no longer supported; variable importance results are always returned if the model.method is set to 'pls' or 'rf'.

Details

Calls pretreat_spectra, format_cv, and train_spectra functions.

Value

list of 5 objects:

'model.list' is a list of trained model objects, one for each pretreatment method specified by the pretreatment argument. Each model is trained with all rows of df.
'summary.model.performance' is a data.frame containing summary statistics across all model training iterations and pretreatments. See below for a description of the summary statistics provided.
'model.performance' is a data.frame containing performance statistics for each iteration of model training separately (see below).
'predictions' is a data.frame containing both reference and predicted values for each test set entry in each iteration of model training.
'importance' is a data.frame containing variable importance results for each wavelength at each iteration of model training. If model.method is not "pls" or "rf", this list item is NULL.

'summary.model.performance' and 'model.performance' data.frames summary statistics include:

Tuned parameters depending on the model algorithm:
- Best.n.comp, the best number of components
- Best.ntree, the best number of trees in an RF model
- Best.mtry, the best number of variables to include at every decision point in an RF model
RMSECV, the root mean squared error of cross-validation
R2cv, the coefficient of multiple determination of cross-validation for PLSR models
RMSEP, the root mean squared error of prediction
R2p, the squared Pearson’s correlation between predicted and observed test set values
RPD, the ratio of standard deviation of observed test set values to RMSEP
RPIQ, the ratio of performance to interquartile difference
CCC, the concordance correlation coefficient
Bias, the average difference between the predicted and observed values
SEP, the standard error of prediction
R2sp, the squared Spearman’s rank correlation between predicted and observed test set values

Author(s)

Jenna Hershberger jmh579@cornell.edu

Examples


library(magrittr)
ikeogu.2017 %>%
  dplyr::rename(reference = DMC.oven,
                unique.id = sample.id) %>%
  dplyr::select(unique.id, reference, dplyr::starts_with("X")) %>%
  na.omit() %>%
  test_spectra(
    train.data = .,
    tune.length = 3,
    num.iterations = 3,
    pretreatment = 1
  )

Train a model based predict reference values with spectral data

Description

Trains spectral prediction models using one of several algorithms and sampling procedures.

Usage

train_spectra(
  df,
  num.iterations,
  test.data = NULL,
  k.folds = 5,
  proportion.train = 0.7,
  tune.length = 50,
  model.method = "pls",
  best.model.metric = "RMSE",
  stratified.sampling = TRUE,
  cv.scheme = NULL,
  trial1 = NULL,
  trial2 = NULL,
  trial3 = NULL,
  split.test = FALSE,
  seed = 1,
  verbose = TRUE,
  save.model = deprecated(),
  rf.variable.importance = deprecated(),
  output.summary = deprecated(),
  return.model = deprecated()
)

Arguments

df

num.iterations

Number of training iterations to perform

test.data

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

best.model.metric

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

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

trial2

trial3

split.test

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.

rf.variable.importance

DEPRECATED rf.variable.importance = FALSE is no longer supported; variable importance results are always returned if the model.method is set to 'pls' or 'rf'.

output.summary

DEPRECATED output.summary = FALSE is no longer supported; a summary of output is always returned alongside the full performance statistics.

return.model

DEPRECATED return.model = FALSE is no longer supported; a trained model object is always returned alongside the full performance statistics and summary.

Value

list of the following:

model is a model object trained with all rows of df.
summary.model.performance is a data.frame with model performance statistics in summary format (2 rows, one with mean and one with standard deviation of all training iterations).
full.model.performance is a data.frame with model performance statistics in long format (number of rows = num.iterations)
predictions is a data.frame containing predicted values for each test set entry at each iteration of model training.
importance is a data.frame that contains variable importance for each wavelength. Only available for model.method options "rf" and "pls".

Included summary statistics:

Tuned parameters depending on the model algorithm:
- Best.n.comp, the best number of components
- Best.ntree, the best number of trees in an RF model
- Best.mtry, the best number of variables to include at every decision point in an RF model
RMSECV, the root mean squared error of cross-validation
R2cv, the coefficient of multiple determination of cross-validation for PLSR models
RMSEP, the root mean squared error of prediction
R2p, the squared Pearson’s correlation between predicted and observed test set values
RPD, the ratio of standard deviation of observed test set values to RMSEP
RPIQ, the ratio of performance to interquartile difference
CCC, the concordance correlation coefficient
Bias, the average difference between the predicted and observed values
SEP, the standard error of prediction
R2sp, the squared Spearman’s rank correlation between predicted and observed test set values

Author(s)

Jenna Hershberger jmh579@cornell.edu

Examples


library(magrittr)
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() %>%
  train_spectra(
    df = .,
    tune.length = 3,
    num.iterations = 3,
    best.model.metric = "RMSE",
    stratified.sampling = TRUE
  ) %>%
  summary()