Problem: for my machine (machine spec mentioned below), installing requirements on this repo, as given in requirements.txt throws the following error.

error: legacy-install-failure

× Encountered error while trying to install package.
╰─> bluepy

note: This is an issue with the package mentioned above, not pip.
hint: See above for output from the failure.

This error is in context with mention of bluepy in requirements.txt.

Machine Specs:

1. miniconda venv for python 3.9.12 running on MacOS Monterey; CPU: Apple M2.
2. miniconda venv for python 3.10.4 running on Ubuntu Jammy Jellyfish; CPU: AMD Ryzen.

Interesting Note: I don't find bluepy being directly imported in the code on the master or the CDT-app-gui branch.

Proposed Solution:

1. Remove bluepy from requirements.txt

Note: This is not a problem with installing PyTranskit itself. It installs pretty gracefully through pip.

In the radoncdt.py file passing the option filter was not working since scikit-image expects the key filter_name.

Tutorial 2 was failing for this reason.

Create a "NS" classifier, as an standalone implementation of the nearest subspace classification method. The "RCDT_NS" and "CDT-NS" classifier can be a subclass of this classifier.

This possibly just needs an edit to reduce the size of the reference signal array alongside the reduction in size of the signal with removed edges.

File "\RCDT_Basic_Tests.py", line 115, in <module>
    Irev = rcdt.inverse(Ihat, temp, nlims)

  File "\pytranskit\optrans\continuous\radoncdt.py", line 123, in inverse
    return self.apply_inverse_map(transport_map, sig0, x1_range)

  File "\pytranskit\optrans\continuous\radoncdt.py", line 235, in apply_inverse_map
    sig1_recon = match_shape2d(sig0, sig1_recon)

  File "\pytranskit\optrans\utils\data_utils.py", line 81, in match_shape2d
    raise ValueError("A is bigger than B: "

ValueError: A is bigger than B: (250, 250) vs (248, 248)