v0.2.6

small bug fix with strides of zero-dimensional arrays

v0.2.5

fiddling some more with precompilation, trying to reduce compilation times

v0.2.4

code simplification and robustness of sreshape

v0.2.3

Some bug fixes; better default arguments for mul! plus in-house fallback implementation

v0.2.2

fallback matrix multiplication is now handled by Strided instead of LinearAlgebra.generic_matmatmul

v0.2.1

• small bug fix in permuting 1D arrays

v0.2.0

• improved implementation of complete reductions, including multithreading
• various fixes
• enhanced test coverage

v0.1.2

fix multithreading strategy to not slice reduction dimensions

v0.1.1

Fix/add initialization step for multithreading capabilities

v0.1.0

First release, fingers crossed!