Philipp Schiele

@jarrodwilcox Can you try if the problem has been resolved with the latest release?

@jarrodwilcox The initial issue of this thread where missing wheels for 3.10, which should be resolved. The error you're seeing now looks similar to #1625, so you could check if...

@SteveDiamond Thanks, I agree that there should not be a reason to deepcopy. Just to prevent introducing subtle bugs here, the following problem: ```py x = cp.Variable(nonneg=True) y = copy.copy(x)...

An alternative would be to override `__copy__` and `__deepcopy__` to fail and point to the `.copy()` function instead.

@SteveDiamond @rileyjmurray Changed to forward `__copy__` to `.copy()` and fail on `__deepcopy__`.

@SteveDiamond are you waiting for another review, or is this PR good to merge?

@Michael-git96 and I have collected a handful of (still mostly synthetic) benchmarks that were posted in various recent issues. This was done as part of the current initiative of writing...

I have also tried the snippet under Ubuntu 20.04, where it worked without issues. The output is indeed puzzling, I have never seen that two consecutive prints would yield different...

@erelsgl Relevant discussions around the topic are here: #1490 / #1493 At the time the `*` import was left it as it came up only as a side effect of...

Hi @datatravelgit ! Looking at the output of pip, is it correct that is building from source rather than using a wheel file? It is correct that the build from...