Monthly Archives: December 2014

First preview of Incanter 2.0 (aka Incanter 1.9.0)

The first preview of the Incanter 2.0 is now available on Clojars as Incanter 1.9.0. This release mostly based on the work done during GSoC 2014 for project “Incanter and core.matrix integration” & includes following big changes (Please look to the wiki for full list of changes):

  • Incanter now uses only core.matrix functions with :vectorz as default implementations. You can use another implementation by adding corresponding dependency to your project file & selecting implementation using the set-current-implementation function;
  • Incanter’s dataset is now based on the Dataset from core.matrix that supports both labeled columns & rows. Dataset is also a matrix now, so matrix functions work on datasets as well;
  • Many functions are now deprecated in favor of corresponding functions from core.matrix (full list of deprecated functions is available on wiki).

Most of these changes are breaking the old API, so please look onto full list of changes that you can find in changelog.

Please, note that this version may have bugs, so please test your code with it & provide feedback. The next versions will provide more functionality from the list of tasks planned for Incanter 2.0, for example, reworking the read-dataset function, etc.

I want to say thank you to all people who participated in this release: Aleksandr Sorokoumov (aka Gerrrr) who did the major part of the work as part of the GSoC 2014; Mike Anderson (aka mikera) for great job on development of core.matrix, vectorz & many other projects & guidance during the GSoC; and all people in the Incanter mailing list who provided the ideas for new release & discussed the ways how to implement this.

Incanter 1.5.6 has been released

I just pushed the Incanter 1.5.6 to Clojars. This is bugfix-only release & includes many changes.

I’m planning to release the first preview of the Incanter 2.0 during the next few days – it will includes the changes made as part of GSoC, including the protocol-based dataset & tight integration with core.matrix.