DoPreprocessing

<code>data.frame</code> 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.

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

test.data

Number or list of numbers 1:13 corresponding to
 desired pretreatment method(s):<ul>
<li>1 = raw data (default)</li>
<li>2 = standard normal variate (SNV)</li>
<li>3 = SNV and first derivative</li>
<li>4 = SNV and second derivative</li>
<li>5 = first derivative</li>
<li>6 = second derivative</li>
<li>7 = Savitzky&lt;U+2013&gt;Golay filter (SG)</li>
<li>8 = SNV and SG</li>
<li>9 = gap segment derivative (window size = 11)</li>
<li>10 = SG and first derivative (window size = 5)</li>
<li>11 = SG and first derivative (window size = 11)</li>
<li>12 = SG and second derivative (window size = 5)</li>
<li>13 = SG and second derivative (window size = 11)</li>
</ul>

preprocessing.method

List of wavelengths represented by each column in
<code>df</code>. Default is 740:1070.

wavelengths

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

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.
Specialized cross-validation schemes are described in detail in Jarqu<c3><ad>n et al. (2017)
<doi:10.3835/plantgenome2016.12.0130>. Example data is from Ikeogu et al. (2017)
<doi:10.1371/journal.pone.0188918>.

DoPreprocessing: Preprocess spectral data according to user-designated method

Description

Usage

Arguments

Value

Examples