Code for simplifications and experiments in Hyper-optimized tensor network contraction

[CSNWEB]

Hello,

thank you for publishing this library, it looks really interesting!

I just read your paper "Hyper-optimized tensor network contraction" and was looking for the code for the simplifications experiments mentioned in that paper, but couldn't find them in this repository. While I am interested in both, I would be most interested in the simplifications. If you could share anything regarding this, that would be really great. It doesn't has to be in a nice shape, I know how easily research code can get messy.

In any case thanks again!

[jcmgray]

Hi @CSNWEB, the tensor network simplification stuff is all in quimb. Specifically:

• full_simplify
• compress_simplify which is more experimental and unpublished..

There isn't any proper examples or documentation I'm afraid - the main thing to note for correctness of the results is that you might need to specify output_inds= if the TN has any hyper indices.