CrystalData: Crystallographic Scattering Data

In GeDS: Geometrically Designed Spline Regression

Crystallographic Scattering Data

Description

This dataset contains crystallographic measurements obtained from a particle accelerator experiment. The measurements correspond to the function F(Q) at various Q values, which are used to analyze the scattering properties of an unknown crystalline material. The dataset is available in two versions based on the precision of the measurements:

• CrystalData10k (lower precision);

• CrystalData300k (higher precision);

The goal of the experiment is to estimate F(Q) from noisy data using a GeDS model, and subsequently compute its Fourier transform to gain some valuable insights into the structure of the material.

Usage

data(CrystalData10k)

data(CrystalData300k)

Format

A data.frame with 1721 observations and 2 variables:

• Q: The scattering vector, measured in inverse angstroms, \text{Å}^{-1};

• FQ: The measured function F(Q), given in arbitrary units (a.u.).

Source

Unpublished data from a controlled particle accelerator experiment.

Examples

## Not run: 
# Load the dataset (choose 10k or 300k version)
data('CrystalData10k')

# Fit a GeDS/GeDSboost model and compare how well the intensity peaks are captured
Gmod <- NGeDS(F_Q ~ f(Q), data = CrystalData10k, phi = 0.999, q = 3)
# for CrystalData300k set int.knots_init = 1, phi = 0.999, q = 4, instead
Gmodboost <- NGeDSboost(F_Q ~ f(Q), data = CrystalData10k, phi = 0.9975, q = 4) 

par(mfrow = c(1,2))
plot(Gmod, n = 2)
plot(Gmodboost, n = 2) 

## End(Not run)

GeDS documentation built on June 30, 2025, 9:07 a.m.