dgasmith

The looped matrix multiply that you show is very hard to optimize for in the general case of einsum. Often the looped GEMM is found permuted such as `kbi,kjb->bij`. In this case, heuristics are needed to determine if GEMM is worth it due to unaligned memory copies.

`optimize=True` is generally best when there are more than two tensors in the expression.

The article was a bit light on hard values, does anyone have good sources? Googling, main pages which come up seem to indicate the impact is only ~15% of total GHG emissions [1][2]. This doesn't seem like much overall for sweeping lifestyle changes, are these values wrong or am I missing something here?

Granted [1] isn't a great source, but hard to find.

[1] https://www.meatinstitute.org/index.php?ht=d/sp/i/47385/pid/...

[2] http://www.fao.org/news/story/en/item/197623/icode/

Note that `opt_einsum` is no longer fully merged with NumPy (and likely will not be) as we now support a wider range of inputs, backends (TensorFlow, Dask, etc), and more powerful optimization algorithms for expressions with hundreds of tensors.

Disclaimer: I am the author of the `opt_einsum` package.

Is there a compare and contrast of this editor against other similar efforts such as Jupyter?