The reason is because `I` doesn't have a size, so you can't `full(I)`. I think the most reasonable thing would be for `I` to have an optional size (and then act like an identity operator of that size, so error on the wrong size multiplications, etc.), and then have sparse(I), Diagonal(I), and full(I) do the sensible things when the size is declared (but nothing if it's not). In which case eye(n) would just make an `I` of size `n`.
On Monday, August 29, 2016 at 9:28:40 AM UTC-7, Tobias Knopp wrote: > > But if it does not makes sense from your point to give a full matrix, why > does it make more sense to return a diagonal matrix? It is still lots of > heap allocated memory. So you replace a large performance trap with a > smaller one. What has been gained? We would still have to teach people the > eye should not be used because there is a faster alternative. > > Am Montag, 29. August 2016 18:04:15 UTC+2 schrieb Chris Rackauckas: >> >> As Julio asks, why should the default be a dense identity matrix? Who >> actually wants to use that? I agree `I` should be in more visible in the >> docs and in most cases it's the better option. But if you actually want a >> matrix as an array instead of just an operator (this is reasonable, because >> maybe you're going to modify it), then a Diagonal matrix is sufficient for >> storing any identity matrix and should be the only thing used in most >> cases. If you really want a dense matrix from it, you could always call >> full(A). But there's almost no reason to have a dense identity matrix, and >> so Julia shouldn't be defaulting to it because it's just a performance trap >> for new users coming from MATLAB. >> >> On Monday, August 29, 2016 at 8:31:13 AM UTC-7, Kristoffer Carlsson wrote: >>> >>> What is the confusion? Use eye(n) if you want a dense identity matrix. >>> Use I if you want something that acts like an identity element. Use >>> Diagonal(ones(n)) if you want a diagonal identity matrix. I see no reason >>> at all why eye should be changed. >>> >>> On Monday, August 29, 2016 at 4:32:34 PM UTC+2, Chris Rackauckas wrote: >>>> >>>> But you don't want a sparse matrix. It would not be an efficient way to >>>> actually use it since sparse matrices have a bit of overhead due to their >>>> table structure. Even better would be a Diagonal since it's just an array >>>> with dispatches to act like a diagonal matrix. But best would be to use >>>> the >>>> UniformScaling operator `I` which acts like eye in cases like A-λI but >>>> isn't actually saving a matrix and doesn't actually require iterating over >>>> two arrays at once. I think that amount of confusion in this thread is a >>>> good indicator that eye should be changed. >>>> >>>> On Monday, August 29, 2016 at 7:25:49 AM UTC-7, Kristoffer Carlsson >>>> wrote: >>>>> >>>>> You mean a sparse matrix? >>>> >>>>
