Philipp Schlegel

Apologies for the late response. This should work: ```Python fig, ax = plt.subplots(figsize=(3,4)) hm = cathead.heatmap(data, ax=ax) ```

My biggest grievance with `mkdocs` is that the HTML produced from Jupyter notebooks (which I use for the tutorials because they can also be tested) just looks atrocious.

I've used the former but the latter also looks familiar - probably tried it out and ran into some issue.

Note to self: I think I like [mdocs-gallery](https://smarie.github.io/mkdocs-gallery/generated/gallery/) - it looks visually clean. It would mean rewriting the Jupyter notebooks to simple Python scripts (using the `#%%` syntax to label...

FYI, preview version here: https://navis-org.github.io/navis/ I managed to solve/work-around most of the problems I encountered along the way - will put details in the PR. There are still a bunch...

Sort of related to what Greg said - although maybe not immediately applicable to Dominik's approach: it may be worth considering performing all topological and/or alignment problem on highly simplified...

Another thought: use difference in Strahler order. That might help disambiguate cases where two neurons densely occupy the same area (e.g. local neurons) but the backbones are in different positions,...

Oh is this the one Alex worked on?

Did you generate a scoring matrix for the Strahler distance?

Technically, the README matches the version on PyPI. I really need to make another release... let me look into if there are any TODOs left.