Saransh Surana

I went through the Klein et al. (2019) paper along with some other references on copulas. Before applying or extending rotations, a few preconditions are needed: 1. Inputs should be...

Thanks, Josef, here’s how I’m planning to build this in a way that fits cleanly with the current copula structure and avoids duplicating existing argument handling. 1. Design of the...

Hi Josef, just to confirm, the line “Would you like to visualize these samples, fit rotated copulas to data, or simulate tail dependence? I can help you build that next.”...

👋 First-time contributor here I tried to replicate this issue and wanted to share what I found. I ran a minimal Gamma regression with and without `var_weights`, and compared the...

Hi @firest11, thanks for the follow-up. I do not have the exact code I used anymore, sorry about that. I can try to recreate it if it would still be...

Thanks for clarifying @josef-pkt. My numbers came from the McCullagh and Nelder convention, while statsmodels follows the SPSS and SAS style with weights folded into the scale. That explains the...

Thanks @josef-pkt , that makes sense. The difference is in the parameterization of weights. McCullagh and Nelder (Ch. 8–9) write the log likelihood for Gamma with weights as ll_i =...

Hi, I’d like to take this issue. My plan is to add a select_list selector that detects columns containing list-like values (lists or tuples). It’ll follow the same structure as...

That sounds like a really neat generalization. Having a has_dtype() selector would make things clearer and backend-specific without extra overhead. I would be happy to help if this gets explored...

I am not sure what the exact problem is since I do not know the details of the code you wrote, but from the error message it looks like LoRA...