- Fix xgboost float precision error
- Configure github actions
- Fix build issue due to dontrun decorator
- Removed from cran, republish.
- Update bm25 function
- Add functions for dot product, svd, normalise matrix
- Add documentation for rcpp functions
- Fixed bugs.
- Introduce rcpp and rewrite base functions in rcpp to create count/tfidf matrix
- Much faster execution with core rcpp functions.
- Deprecate
parallelargument, since turned out to be slower than c++. - Fixed bugs.
- Fix bugs in CountVectorizer & TfidfVectorizer
- Add ngram_range feature
- Add more documentation on vectorizers
- glmnet changed its api. Fix Failing tests. Temporary fix by using "donttest" on its examples. Updates package documentation to support R6 classes.
- Fixed svm error. Replaced liquidSVM with e1071 R package.
- Moved some imports to suggests
- Added new trainers: SVM, NaiveBayes
- Renamed gridsearch, randomsearch trainers to CV
- Fixed documentation
- Added a
NEWS.mdfile to track changes to the package. - Fixed travis-ci integration