The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.

squat 0.4.0

New features

• Adaptation to new API of fdacluster.
• Added rotation invariant elastic distance.
• Added no alignment option in distance matrix computation via SRVF.
• Added roll-pitch-yaw representation.
• Added format()/print() pair correctly.

Minor improvements and bug fixes

• Remove dependency to purrr in favor of base R *apply() family to reduce dependency trail.
• Replace furrr with future.apply for consistency with the removal of purrr.
• Update roxygen version.
• Fix bug in predict from PCA with new data. Properly compute L2 inner product.
• Fix bug in predict from prcomp object.

squat 0.3.0

New features

• Added S3 specialization of the stats::predict() function for prcomp_qts objects.
• Added function qts2aats() which allows to switch from quaternion to axis-angle representation of rotations.
• Added usual operations +, -, * and inverse_qts() for quaternion time series using the Eigen library.

Small improvements

• Make sure quaternion geodesic mean is not flipped.
• Fix issues in PCA:
  • avoid numerical overflows due to bad choice of k in gam();
  • improved documentation;
  • Use same number of basis functions in uni- and multivariate decompositions.
• Updated GHA scripts and README.
• Fix bug related to Rcpp following RcppCore/Rcpp#1287.

squat 0.2.1

• Add use_fence robustification option;
• Adapt to changes in fdacluster;
• Properly compute tangent spaces along mean QTS;
• Update News section of website.

squat 0.2.0

Major features:

• Added hierarichal clustering;
• Added DBSCAN clustering;
• Added distance matrix computation.

Minor improvements:

• Adapted code to match new API in fdacluster package.

squat 0.1.0

Major statistical features

• A first API proposal with a class qts and a class qts_sample for which a number of methods are properly implemented.
• Available statistical methods for QTS samples:
  • random generation according to the Gaussian functional model via rnorm_qts(),
  • scale(),
  • mean(),
  • median(),
  • distance matrix computation via distDTW() (i.e. for now we use the dynamic time warping),
  • tangent principal component analysis via prcomp(),
  • k-means with optional alignment via kmeans().
• Added multiple ways of displaying samples of QTS.
• Added two example datasets.

Improvements

• Make all functions applicable to a single QTS also applicable to QTS samples, with appropriate class for the output.
• Enable as_qts_sample() to generate a QTS sample of size 1 from a single QTS as input argument.
• Rename change_points argument to the plot.qts() function to better reflect its flexibility.
• Added subset operator for QTS sample objects.
• Added append S3 method for QTS sample objects.
• Added hemispherize() function to remove any discontinuities in QTS due to quaternion flips.
• Any parallelization computation is now handled using the futureverse principles and, in particular, implemented through the use of the furrr package.

squat 0.0.1

• Added a NEWS.md file to track changes to the package.
• Initial version.

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.