Alexey Stukalov
Alexey Stukalov
This is an attempt to implement some parallelization. Of course, the most straightforward way would be to parallelize fitness calculation in `rank_by_fitness!()`, but since Julia doesn't have threads [yet], it...
Just to have it in the list. [Rtnse](https://github.com/jkrijthe/Rtsne) could be run in "exact" (what TSne.jl has now) and "approximate" modes. In "approximate" mode it optimizes the calculation of gradient in...
`tsne(X)` treats `X` rows as observations (points) and columns as their features (dimentions). I guess it's because the original implementation is written in Python/NumPy, which uses row-major order. However, Julia...
This PR * simplifies a bit the wrapper generation logic * some unnecessary "type piracy" was removed (`A*B` calls `mul!(C,A,B)` already) * streamlines testing, so that all 4 numeric types...
Allows specifying custom paths that would be used for MKL injection into *sysimg.jl*. The paths default to the current `["@v#.#", "@"]`, but could be explicitly specified when calling `enable_mkl_startup()`, or...
I'm trying to display some block-matrix-like plot, and there are a few regressions when converting from ggplot to ggplotly conversion (ranked by severity): * degenerated (1 value) discrete axes are...
*shinystan* is a quite cumbersome dependency, but its use is isolated to `launch_shinystan.brmsfit()`. Would be nice to make it optional.
This PR reenables the support for *PriorContext* in optimization, which was disabled by #2022. While it is not directly supported by the current *optim_problem()* interface (only MLE/MAP are supported, although...
This is a subset of #30 excluding the `Matrix * SparseMatrix`: * adds support for MKL 2023.x * simplifies the MKLSparse wrappers by introducing `mkl_call(funcname, T, args...)` generated function that...
When I look at the `sem(..., group=..., group.equal="loadings", rotation="geomin")` results -- I observe that the loadings for the groups are different, while the unrotated loadings are the same. Also, in...