Jared Huling

I'm having this issue as well, but >runtime error: member call on address ... which does not point to an object of type 'BigMatrix' ...: note: object is of type...

@YaohuiZeng Did you ever get this issue resolved? If so what was the fix?

In general, with glms, the estimated variance-covariance matrix comes from $(X'WX)^{-1}$ from the last iteration of IRLS. Since the fastglm package offers different methods for solving the weighted least squares...

Could you expand on what you mean by this?

I've tinkered with a basic implementation of this via bigmemory in [src/glm_big.h](https://github.com/jaredhuling/fastglm/blob/master/src/glm_big.h) but I have not had time to carefully check whether it truly works as it should

Sure, I should be able to do that. It might be a couple of weeks, as I've got some other things going on. Just so you know, the paper from...

I added sparse matrix support (namely SparseMatrix format from the Matrix package). Note that while this is faster for sparse matrices, the returned inverse will in general not be sparse....

It seems that this algorithm tends to given incorrect results more frequently for sparse matrices. I don't recommend using it. I will replace it when I find a good alternative

sure, there are definitely cases where the inverse of a sparse matrix is sparse, but that's not true for many cases (most I've come across) ``` rout library(rfunctions) #create matrix...

For that size geninv() as it currently stands won't be very memory-efficient. I'm working on a way to fix that