The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
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[@]metanet.ch.

traj 3.0.0

• Substantial modifications to the measures.

• The clustering step now relies of a version of spectral clustering

• The main function are trajMeasures (computes the measures) and trajClusters (finds the clusters), with trajReduce (finds a representative subset of measures) being accessory.

• The plotting functions that trajClusters can be passed into are now plot, scatterplots and CVIplot.

traj 2.2.1

• Added a new measure, “m5: slope of linear model”.

• Spit the “plot” function into three: plot, plot.scatter and plot.crit.

• Improved the presentation of the scatter plots.

• Made Step1Measures more lenient with how the input data is formatted.

traj 2.2.0

• Added k-medoids as the default clustering algorithm.
• Added the Calinski-Harabasz index as the default criterion for determining the optimal number of clusters.

traj 2.1.0

• Makes substantial changes to the list of measures.
• In trajdata, the group of size 30 is made up of quadratic (instead of linear) curves.
• Introduces the vignette “Using the traj package”.

traj 2.0.1

• Fixes minor bugs and improves presentation.

traj 2.0.0

• Changes the way the measures are computed in order to be better suited to missing values and unequally spaced observation times.
• Allows for better control over the choice of a “midpoint”.
• Implements a capping procedure to automatically handle outliers. Measures of the form 0/0 are set to 1.
• Fixes an important bug in the implementation of step 2.
• The criterion of Tibshirani et al. based on the GAP statistic is the new default for choosing the number of clusters.
• Allows more control over the parameters of kmeans.
• The outputs of summary, print and plot are more detailed.

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.