Hi. > > [...] > > > > [1] The inconsistencies that led to the deprecation will have no bearing > > on usage within the optimizers. > > [2] The latter option seems likely to entail a fair amount of work to > > improve the performance of "OpenMapRealMatrix" (which is quite poor > > for some operations, e.g. "multiply"). > > > > I have never been happy with deprecating the sparse implementations of > vectors and matrices. I also think we should "undeprecate" them and work on > them, with some help! > Regarding [1], my choice would be to point at the problematic cases in the > Javadoc (and possibly the user's guide), and live with them. > Regarding [2], would there be any performance gain if we designed an > immutable sparse vector/matrix. I guess it would. The question is: do we > need these objects to be mutable?
I don't think so. And certainly not for the usage in the optimizers. Anyways, I assume that it would be quite useful to have a new matrix hierarchy (with more types than we have now, and appropriately optimized for storage and/or speed). And a good starting point would be to assume immutability. Best, Gilles --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org For additional commands, e-mail: dev-h...@commons.apache.org
