Here's the development roadmap. In the short term, we should focus on:
In the long term we should consider several generalizations/specializations:
Also, we should think of a cool name. We should convey a message that this solver targets structured sparse-dense systems
Here's the development roadmap. In the short term, we should focus on:
factorize!andsolve!doesn't allocate memoryIn the long term we should consider several generalizations/specializations:
Also, we should think of a cool name. We should convey a message that this solver targets structured sparse-dense systems