Changes in version 0.5.2 (2025-04-25)                  

Bug fixes

  - Update tests to pass CRAN check with Intel MKL.

                 Changes in version 0.5.1 (2025-04-24)                  

Bug fixes

  - Add workaround for a compilation error when clang 20 is used (#65).

                 Changes in version 0.5.0 (2025-04-22)                  

New features and improvements

  - Enable simil() nad dist() to perform masked similarity/distance
    computation. If a pattern matrix is given via the mask argument, it
    computes scores for selected pairs of rows or columns. Pattern
    matrices can be created using newly added mask() function.
  - Calculate "dice" and "edice" using Armadillo's linear algebra
    functions in simil(). As a result, computation of these scores
    became as fast as "cosine", "correlation" and "euclidean".
  - Add crossprod() and tcrossprod() using the same infrastructure as
    simil() and dist().

                        Changes in version 0.4.2                        

New features and improvements

  - Reduce the overhead for dense similarity matrices by improving
    rounding numbers and conversion to Rcpp vectors.
  - Return a dense matrix if sparse = FALSE to save space in RAM.

                 Changes in version 0.4.1 (2024-04-07)                  

Bug fixes

  - Make detection of Intel oneAPI Threads Building Blocks (TBB) library
    more reliable.

                 Changes in version 0.4.0 (2024-04-05)                  

New features and improvements

  - Use more recent Intel oneAPI Threads Building Blocks (TBB) library
    to improve the stability in parallel computing.
  - Add options(proxyC.threads) to control the number of threads in
    parallel computing (but RCPP_PARALLEL_NUM_THREADS still has effect).

New system requirements

  - The RcppParallel package is no longer required as the TBB library in
    the operating system (Linux and MacOS) or Rtools (Windows) is used.
  - Linux and MacOS must have the TBB library to enable parallel
    computing before installing this package from the source.

                 Changes in version 0.3.4 (2023-10-25)                  

New features and improvements

  - Add "fjaccard" to simil() for Fuzzy Jaccard similarity (#42).

                 Changes in version 0.3.3 (2022-10-06)                  

New features and improvements

  - Explicitly setting use_nan = FALSE will suppress warnings in simil()
    and dist().
  - Add vignettes to explain how the similarity and distance measures
    are computed.

                 Changes in version 0.3.2 (2022-09-05)                  

Bug fixes

  - Make further changes for Matrix v1.4-2.

                 Changes in version 0.3.1 (2022-08-22)                  

New features and improvements

  - Add "jensen" to dist() for Jensen-Shannon divergence as a symmetric
    version of Kullback-Leibler divergence.
  - Change how x and y are coerced to dgCMatrix for Matrix v1.4-2.

                 Changes in version 0.3.0 (2022-08-05)                  

New features and improvements

  - Add "jeffreys" to dist() for Jeffreys divergence. It is a symmetric
    version of Kullback-Leibler divergence (#31).

                 Changes in version 0.2.4 (2021-12-10)                  

New features and improvements

  - rowSds(), colSds(), rowZeros() and colZeros() return row or column
    names. They also work with both dense and sparse matrices (#28).

                 Changes in version 0.2.3 (2021-11-16)                  

Bug fixes

  - Change "hamman" to "hamann" in simil() to correct misspelling (#26).

                 Changes in version 0.2.2 (2021-10-27)                  

New features

  - simil() and dist() work with both dense and sparse matrices.
  - use_nan = TRUE can be used not only for correlation but for all the
    distance and similarity measures.

                 Changes in version 0.2.1 (2021-09-02)                  

New features

  - Computing the correlation similarity on vectors with a standard
    deviation will generate a zero correlation and a warning. The
    warning can be turned off by setting use_nan = TRUE, in which case
    the computed correlation similarity will be NaN instead (#21).

Bug fixes

  - Fixed infinite values being returned by the correlation similarity
    (#20).

                 Changes in version 0.2.0 (2021-05-11)                  

New features

  - Added a diag argument to compute similarity/distance only for
    corresponding rows or columns (#13).
  - Added a smooth parameter to chisquared and kullback leibler
    distances to solve negative values in sparse matrices (#15).
  - Added the hamming distance (#18)

Bug fixes

  - Fixed the chi-squared distance to match stats::chisq.test() (#14).
  - Fixed a bug in pairwise similarity/distance computation when drop0 =
    TRUE (#17).

                 Changes in version 0.1.5 (2019-07-21)                  

New features

  - Add the drop0 argument to address the floating point precision issue
    (#10).

Bug fixes

  - The digit argument is now passed to dist() (#11).

                 Changes in version 0.1.4 (2019-06-04)                  

New features

  - Add rowSds(), colSds(), rowZeros() and colZeros() (#9).

                 Changes in version 0.1.3 (2019-04-21)                  

Bug fixes

  - No longer assumes symmetry of resulting matrix when x != y (#4).

New features

  - Add the digits argument to correct rounding errors in C++ (#5).