hadware

Woah great work @lumbric ! That's actually quite an hacky yet elegant way to make this work. I'll try to see if my unit tests fail or not when I'm...

@SteveDiamond nice! However, in a previous message, you said that > The issue is that the solutions from ECOS_BB were often totally wrong. Are these issues solved?

Sidenote: we may be able to pretrain a `MALE`/`FEMALE`/`CHILD`/`KEYCHILD` model using our internal data, but this depends on the severely restrictive (and rightfully so) nature of the data. This could...

In my case, it's this one: https://github.com/coin-or/python-mip/blob/2f23cec23f886847738f7fe2bbe9064e2ed86919/mip/entities.py#L591 (since i'm multiplying a `LinExprTensor` with a numpy array), but I'm guessing it could be any of the instance checks from the `Variable`...

The A matrix was relatively sparse (I'm [sending](https://github.com/coin-or/python-mip/files/5330733/arrays.zip) the two matrices as attachment). I kind of looked into the cvxpy internals to see how their way of doing things could...

> Also, as an alternative use case. It would be nice if the tokens field can be a list instead. That way we can define the proper order of applying...

I agree with this. This is also linked to https://github.com/soundata/soundata/issues/154 and should be solved.

@jpsnyder : :thought_balloon: Would you confirm that a synthetic version of the human mind can produce suggestions that are valid in the context of software development?

That is also exactly what I did to get a horizontal graph

Sure