Functional ransac

[yjmantilla]

PR Description

Made ransac methods into their function versions.
Then moved the functions into a standalone module called ransac.
Some changes were made to the parameters of the methods as a result.

Since I moved the stuff I dont know what to put in the RST for old entrances (ie :meth:NoisyChannels.run_ransac) and so on. This is now in :func:ransac.run_ransac. But we can move everything back to find_noisy_channels if that is a better way.

I also made the channel_correlations an output of the find_bad_by_ransac function for easier debugging, but I dont know if this is the best approach, so suggestions are welcome. It could also be a class that holds the end results (no mutable attributes), or maybe a dictionary.

Merge Checklist

To merge your PR we need to first take the following points into account:

• the PR has been reviewed and all comments are resolved
• all CI checks pass
• (if applicable): the PR description includes the phrase closes #<issue-number> to automatically close an issue
• (if applicable): bug fixes, new features, or API changes are documented in whats_new.rst

[sappelhoff]

@a-hurst can you offer your opinion here as well, please?

[yjmantilla]

Note that the idea of this code is to be able to do #36 , not to make the ultimate ransac function that in theory should arise from the comparison we make.

[sappelhoff]

I suggest to make all functions in ransac.py except find_bad_by_ransac private. That means putting an underscore _ in front of the function names.

Furthermore, the new module needs to be documented in api.rst, please only document find_bad_by_ransac there --> you can copy (or get inspired by) the syntax from the already existing sections.

Lastly, can you please add ransac.find_bad_by_ransac here:

pyprep/pyprep/__init__.py

Lines 1 to 8 in dd618a3

	"""initialize pyprep."""
	from pyprep.find_noisy_channels import NoisyChannels # noqa: F401
	from pyprep.prep_pipeline import PrepPipeline # noqa: F401
	from ._version import get_versions
	__version__ = get_versions()["version"]
	del get_versions

for example:

 import ransac  # noqa: F401 

So that users can do from pyprep import ransac and then call ransac.find_bad_by_ransac

(actually this might be possible also without editing init.py, please feel free to check prior to changing it)

[codecov-io]

Codecov Report

Merging #51 (4d915bd) into master (bfb2ab1) will increase coverage by 0.03%.
The diff coverage is 97.05%.

@@            Coverage Diff             @@
##           master      #51      +/-   ##
==========================================
+ Coverage   96.76%   96.80%   +0.03%     
==========================================
  Files           5        6       +1     
  Lines         526      532       +6     
==========================================
+ Hits          509      515       +6     
  Misses         17       17              

Impacted Files	Coverage Δ
pyprep/ransac.py	96.96% <96.96%> (ø)
pyprep/find_noisy_channels.py	96.34% <100.00%> (-0.16%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update bfb2ab1...4d915bd. Read the comment docs.

[sappelhoff]

@yjmantilla I pushed several commits to master to fix the documentation links and improve a few maintenance features.

I now rebased your PR on top of these changes, and also added the new module to the API docs (see 67a4d2c)

before continuing to work you will have to:

git pull
git reset --hard upstream/master
git pull

then it should tell you that everything is up to date and fine. (assuming you have a git remote "upstream" configured to point to sappelhoff/pyprep)

I just hope you didn't have non-committed changes in your working directory :-)

[yjmantilla]

I have not forgotten this yet, just that now I'm doing the final projects of the semester so haven't got time to work on this until around the 25 of January.

Cheers! And Happy New Year!

[sappelhoff]

Happy new year to you too @yjmantilla!

[sappelhoff]

I finally got the links within the docs working :-)

working beautifully now: https://pyprep.readthedocs.io/en/latest/generated/find_noisy_channels.NoisyChannels.html#find_noisy_channels.NoisyChannels

try to hover over the types like "float" or "np.random.RandomState" --> these are now links that will lead you to the docs of Python3 or Numpy respectively.

I rebased this PR on my changes, you should be able to just continue where you left off once you find time 👍

[yjmantilla]

Sorry for the lateness (rough year start). Everything should be in place now. After this is merged next step is to make the example file comparing autoreject's and pyprep's versions of ransac.

[sappelhoff]

Thanks @yjmantilla hope nothing bad happend