Peter H. Li

In general it might be nice if the user can specify names of codegen functions more, but since the different codegens do fairly different things (solvable Python code, compilable C...

Sounds right, but in general it's nice for the names of the functions to match somehow the names of the files. Maybe codegen can be tweaked so that the function...

These are sharded files, so '@1000' means the dataset is split into 1000 different files. You then also see this in the individual file names '00000-of-01000' etc. The training examples...

The max200000 and skip50 in the name refer to hyperparameters chosen in the extraction of the example pairs. I'm pretty sure max200k means that the maximum pair distance sampled is...

> So it also means that these 1000 files do not correspond to 1000 segments, but rather all the sampled pairs are combined and averaged into 1000 files. You mentioned...

We used the existing skeletonization of the h01 c3 segmentation. This was done via the improved TEASAR method implemented in the [kimimaro](https://github.com/seung-lab/kimimaro) package. Once you have skeletons, you can sample...