jgostick

In the ``COSTALD_compressed`` method, and perhaps others, there is an argument ``Vs`` for saturation volume. It was my impression that the suffix ``s`` on an argument name meant *plural* as...

Replace random steps with mean free path?

4

@TomTranter Regarding the idea of doing Knudsen tortuosity with this package, how hard would it be to replace the random step with a predetermined step based on a mean free...

Better parallelization using Dask?

2

We've had some great success with dask for simple single-machine parallelization. I think the current version uses something a bit problematic right, requiring the script to be in a __main__...

The arg `saturation` should be changed to `satn` for consistency, and the docstring does not describe what is returned on the Results object

Use keyword `method` instead of `mode`

3

I have noticed that some numpy/scipy functions have (started) using this. I think I like it and it's slightly more descriptive. Is it worth making the change?

Enforcing this convention makes it obvious which attributes are images.

Using `conn` consistently throughout

2

This may need a bit of thought...currently we accept `conns` in some places to mean an structuring element, and in other places we have `conns=4|8` in 2D and `conns=6|26` 3D....

remove `return_spheres` arg from `random_spheres`

1

This is to be consistent with the the pseudo packing functions. OR, maybe we should add the `return_spheres` arg to the pseudo packing functions?

Replace `overlay` with `insert_shape_at_point`

2

The overlay function should accept a mode, which can be one of `'add', 'preserve', 'overwrite'`. It's currently only doing `'add'`.