Ben Pedigo
Ben Pedigo
@daxpryce do you want to re-review?
sure to fixtures or any other computationally smart stuff what i meant was lets literally make them less computationally burdensome (fewer samples, smaller data, etc.)
just as an example https://github.com/microsoft/graspologic/blob/bfcfde4bbb9b4ac251596eb9ebcb3b9496f3446e/tests/test_sims.py#L906
have discussed some examples where regularized LSE worked better (at least for me) with @daxpryce and @Nyecarr - I feel like regularized is what I'd recommend anyone using. Also, `form='I-DAD'`...
`None` (means regularizer is set to the mean degree)
@daxpryce @Nyecarr just wondering if there is anyone on your side working on this at the moment?
## Tracking progress ### Main code - [x] new solver class - [x] bilateral - [x] multilayer - [x] seeds - [x] similarity - [x] option to turn off numba/use...
## Benchmarking - [ ] Timing - [ ] new vs old - [ ] sparse vs non-sparse - [ ] parallel vs non-parallel - [ ] Performance - [...
@sagarhowal that's great, we'd appreciate any help/input. And no, we didn't have preexisting ideas. But I like your doodle! I'm wondering if we deleted the edges that are outside of...
i do note that we've thus far mostly been using graspologic (lower case) instead of Graspologic - personally I don't care much, and I think the G above looks cool....