gaussplotR 0.2.5
- Citation information updated
- Ready for updating CRAN
gaussplotR 0.2.4
- Post-peer review at the Journal of Open Source Software.
print()
and summary()
on objects that are
output from fit_gaussian_2D()
now round parameter estimates
to two decimal places.
- R version requirement downgraded to R >= 3.3.0
gaussplotR 0.2.2
- Newest functions enhance the automation of comparing among various
Gaussian fit models.
- The
autofit_gaussian_2D()
function can be used to find
the best-fitting model for a given data set.
- The
compare_gaussian_fits()
function compares models
via criteria such as rmse or rss.
- A
characterize_gaussian_fits()
analyzes the orientation
and partial correlations of Gaussian data. Features include computation
of partial correlations between response variables and independent and
diagonally-tuned predictions, along with Z-difference scoring.
gaussplotR 0.2.0
- Added a
fit_gaussian_2D()
function to apply any of
several available methods to fit a 2D-Gaussian to supplied data
predict_gaussian_2D()
has been enhanced to allow any
method to be used in fit_gaussian_2D()
and then fed into
the predict function
gaussplotR 0.1.6
- Added a
compute_gaussian_volume()
function to compute
the volume under a given 2D-Gaussian
gaussplotR 0.1.4
- Miscellaneous formatting fixes for CRAN checks. No changes to
functions.
gaussplotR 0.1.3
- Added a
NEWS.md
file to track changes to the
package.
- Package now seems to be ready for submission to CRAN