Accuracy issues with batches

[PaulR-S]

Hi, I am using the qpth solver to solve many qp's (i.e. core price computations) as batches in parallel. Now, comparing the prices (resulting variables) with the ones I receive from gurobi shows that the ones from qpth differ quite a bit (see below). It seems like this is more likely to happen with a batch size > 1.

I tried to use float and double as well as cpu and gpu. The result of qpth is always the same (or at least very similar).
I also compared the result from cvxpylayer to gurobi and they are very close.

This is what I get when setting verbose to True

Do you know why the accuracy is so bad and how it can be improved? I'm using pytorch 1.3.0., qpth 0.0.15, gurobi 8.1.1. Thanks.

[bamos]

Hmm, for your problem here, how is the speed in comparison to cvxpylayers? I've been thinking about deprecating this repo and pointing everybody to cvxpylayers

[PaulR-S]

Cvxpylayers allows batches, however, not on the GPU but CPU only, correct? For the setting I ran, cvxpylayers was significantly slower, in fact it was slower than parallelizing the batch over multiple CPU's and solving each with Gurobi.

[bamos]

Cvxpylayers allows batches, however, not on the GPU but CPU only, correct? For the setting I ran, cvxpylayers was significantly slower, in fact it was slower than parallelizing the batch over multiple CPU's and solving each with Gurobi.

Hmm, I think we if we add in support for other solvers to cvxpylayers we can outperform qpth by using OSQP, for example. I've added an issue to cvxpylayers you can use to track our development on here -- and if you have some bandwidth you are welcome to help out with the development there to get this in faster: https://github.com/cvxgrp/cvxpylayers/issues/28

\cc @bstellato @sbarratt @akshayka