Zulip Chat Archive

Stream: Is there code for X?

Topic: neg as a linear map


Sébastien Gouëzel (Apr 14 2024 at 19:23):

I'm still lost in the maze of naming conventions for linear maps. What is the name of the linear map mapping x to -x in a module? I've found docs#IsLinearMap.isLinearMap_neg, which is almost this but not exactly. I'm missing a tool for which I would give the toFun field, and which would browse through all the definitions of linear maps we have to find me a map with this field.

Yury G. Kudryashov (Apr 14 2024 at 19:24):

Does -id work?

Yury G. Kudryashov (Apr 14 2024 at 19:25):

https://leanprover-community.github.io/mathlib4_docs/Mathlib/Algebra/Module/LinearMap/Basic.html#LinearMap.instNegLinearMapToAddCommMonoid

Eric Wieser (Apr 14 2024 at 21:32):

docs#LinearEquiv.neg ?

Yury G. Kudryashov (Apr 14 2024 at 21:59):

Should we have Neg instance on LinearEquiv instead of LinearEquiv.neg?

Eric Wieser (Apr 15 2024 at 07:21):

@loogle LinearEquiv, Neg

loogle (Apr 15 2024 at 07:21):

:shrug: nothing found

Eric Wieser (Apr 15 2024 at 07:22):

Maybe? Though we can't do the same for docs#AddEquiv.neg'

Eric Wieser (Apr 15 2024 at 07:22):

So it would leave us with a primed name and no unprimed counterpart

Sébastien Gouëzel (Apr 15 2024 at 07:27):

For the record, in the end I used LinearIsometryEquiv.neg (which I found by redefining it and then getting a complaint from Lean that there was a name collision :-)

Yury G. Kudryashov (Apr 16 2024 at 01:08):

BTW, the docstring for docs#AddEquiv.neg' mentions AddEquiv.inv which doesn't exist.


Last updated: May 02 2025 at 03:31 UTC