Biscuit-dough NIR spectra
=========================

Source
------
The ``cookie`` data set of the R package ppls 2.0.0 (CRAN), file data/cookie.rda.
The R packages groc 1.0.10 and seedCCA 3.1 (CRAN) carry the same values.

References
----------
Osborne, B. G., Fearn, T., Miller, A. R. and Douglas, S. *Application of near infrared
reflectance spectroscopy to the compositional analysis of biscuits and biscuit doughs.*
Journal of the Science of Food and Agriculture, 35(1), 99-105 (1984).
https://doi.org/10.1002/jsfa.2740350116

Brown, P. J., Fearn, T. and Vannucci, M. *Bayesian wavelet regression on curves with
application to a spectroscopic calibration problem.* Journal of the American Statistical
Association, 96(454), 398-408 (2001). https://doi.org/10.1198/016214501753168118

Contents
--------
biscuit.csv: 72 unbaked biscuit doughs, one row each. ``sample`` numbers them 1 to 72 in
the upstream order; ``set`` is ``calibration`` for the first 40 and ``validation`` for the
other 32; ``fat``, ``sucrose``, ``dry_flour`` and ``water`` are the percentages of each
ingredient; the 700 columns ``1100`` to ``2498`` are the NIR reflectance spectrum, one
column per wavelength in nm, at 2 nm intervals. The values are the upstream values,
unchanged; only the column names of the spectrum are the wavelengths instead of 1 to 700.

Terms
-----
The values are measurements reported by Osborne et al. (1984): facts, not a creative work.
The MIT licence of process-improve covers its code, not this file, and claims nothing over
the measurements. Cite Osborne et al. (1984) when you use them.

The R packages ppls, groc and seedCCA are distributed under the GNU General Public License,
version 2 or later. That is the licence of those packages, whose authors did not make the
measurements. This file is kept separate from the code that reads it, so it is an aggregate
in the sense of the GPL and does not change the licence of process-improve.
