William Zijie Zhang

Yes, I'd be happy to make PR. I was wondering also, how can we adjust the suite-sparse settings? For example, we are considering defaulting to using csc matrices instead of...

Hey thanks for bringing this up Erik, I think it would be a great addition. I also would prefer ``.new(order="columwise")`` but I think all three options are equally suitable for...

Hey @eriknw , thanks for your explanations. If I understand correctly, the first PR is just there to add an optional argument ``order`` to the new() constructor for various Matrix...

Hello @eriknw , sorry for the delay on this, I have been finishing up some details for my GSoC project. I tried to implement the orientation today, however, it is...

Ok, I noticed a slight issue with the changes in this PR. ``np.broadcast_shapes`` is a relatively new API function (see image below) and hence fails the builds for Python 3.9...

This PR looks good on my end. Any feedback, questions or suggestions are very welcome. Ping for reviews: @PTNobel @SteveDiamond @phschiele (when you are back from Mexico 😅 ). Also...

@PTNobel I propose to update this PR to remove all the failing tests with DQCP. Then we can open another PR specifically for the remaining tests as well as drop...

@PTNobel a few tests are still failing because of precision errors (see below). I tried looking in the clarabel [docs](https://clarabel.org/stable/api_settings/#Clarabel.Settings) but I'm still not sure what solver setting to pass...

@PTNobel I fixed this PR by removing the precision error tests mentioned above. I think the failing benchmark (one of them has a "significant" change) is really weird and definitely...

Hello @purva-thakre thanks for raising this issue! Would you be open to making a PR to resolve this? I will let you know if I find more type annotations that...