Vincent Barrielle
Vincent Barrielle
As mentioned in https://github.com/vbarrielle/sprs/issues/278#issuecomment-820984278, I'm currently lacking the time to work on this crate, and I'm unsure when I'll have the time and motivation again. I think it's probably a...
More tests stressing out the limits of small integer types should be added. Existing tests could be converted too.
.diagonal would allocate a sparse vector while .diag_view could allocate a lazy diagonal proxy, trading time for memory
Requiring conversion before the product can be really slow. This requires setting up some benchmarks to measure the concrete effect.
As explained [here](https://users.rust-lang.org/t/how-to-zip-two-slices-efficiently/2048), using zip to iterate on members with the same length is inefficient bacause the compiler won't be able to remove both bounds checkings. Some tricks exist for...
Right now this product is either available as a special case of the sparse/dense matrix multiplication, or as a free function that only takes slices as input. We need to...
Have an array view data structure, and an associated trait enabling to take an array view into a matrix. ArrayView implementation of Mul will perform a scalar multiplication instead of...
Since the code is mostly generic, complex types should be supported, but this needs testing. Some complex-only functions should be added.
The rhs could be a dense matrix, or maybe an iterator of rhs. We might want to abstract over these cases.
We need to prevent overflow, if possible by detecting the possible overflows at matrix creation.