Eric Wieser

Again, I'm finding it somewhat hard to follow your comment. Are you talking about using Unicode in the source code, as the names of the basis blades? Or suggesting that...

I'll emphasize that I don't think you want to do this, but if you _did_ you can in `master` now use: ```python import clifford as cf layout = cf.Layout([1, 1,...

Could you add an entry to https://github.com/pygae/clifford/blob/master/docs/predefined-algebras.rst too?

There was a discussion on our slack about what to name the algebra, we should check if that came to a conclusion before merging.

This happens in CGA too: ```python >>> from clifford.g2c import * >>> t = up(e1) | (up(e1)^up(e2)^up(-e1)) # tangent blade >>> t*t 0 >>> t.isBlade() clifford\_multivector.py:264: RuntimeWarning: divide by zero...

Whoops, good catch

Did some more reading, your changes look obviously correct to me. Can you add a test in `test_cga.py` both with and without einf? ~~, and fix the conflicts? (should be...

20% less time spent on multiplication, 30% less time on addition. Possibly a little less, since the benchmark in that PR folds in another change too. But remember, the question...

That's one way of looking at it, We can get the performance boost either by: 1. Adding `copy=False` on 75 lines of our code, and telling users to do the...

CI failures can be ignored. Thanks for the PR, I think I like the `pinv` name better than your previous suggestion anyway.