Flexible Linear algebra library working in arbitrary dimension
This crate propose an alternative strongly inspired from nalgebra that aims to reduce the boilerplate its mentor has for working with generics, while keeping its flexiblity. The result is as much simpler and comprehensive library with a shorter codebase
It mainly features the [Matrix] struct, which represent a dense matrix with any strided memory. And providing common matrix operations as its methods.
Of course this crate address both the case of small-dimension objects better used as static-allocated objects and the case of big-dimension arrays usually handled with dynamic allocation
no-std
feature, but not started yet- as feature-complete as nalgebra and numpy
- as fast as size-specific libraries
At contrary to nalgebra, Matrix
is not a one-size-fit-all struct with tons of generic arguments. It is a wrapper type that can wrap any array storage type (even not from flexalgebra). The type and shape parameters only depend on the array storage which can be user provided.
This means any user-type can be used as a matrix compatible with the rest of the flexalgebra types