Chris Barnes

Great, looking forward to using them! Another possible minor thing to look at is functions which take an `"auto"` string argument, which is different to `None`, and is also different...

This is in 3.10 and may help out type checkers for functions which work on multiple different types: https://www.python.org/dev/peps/pep-0647/

I suspect it'll work with any python once there have been a couple more `mypy` releases, but they'll have to be defined as strings for older pythons. The way backporting...

I think multiprocessing can misbehave if there's more than one layer of concurrency - I think it depends on how everything is implemented (and can be platform-dependent too because the...

Here was my prototype: https://github.com/clbarnes/transformnd (no tests or anything so it may not work). As you can see it's quite cut down, really targeted at being light and generic. navis'...

> facepalm I had actually been looking at your repos to check whether you had already made a start but must have missed this one. Only pushed it today! >...

Yeah, sounds like an extra function `strahler_summary`, which can use a neuron already populated strahler values or populates them if necessary.

Maybe the node IDs at the root and leaf of each segment?

Could we use a [SharedMemoryManager](https://docs.python.org/3/library/multiprocessing.shared_memory.html#multiprocessing.managers.SharedMemoryManager) to handle the cleanup? The context manager form should make it fairly convenient. Doing a single copy of the shared numpy array at the end...

If you don't have the memory to copy the score matrix in memory then you can just do whatever you want to it inside the context manager, I suppose. If...