josephcappadona / Twine-Technical-Project

Technical project built as part of the interview process for Twine Labs

Instructions

To use,

python twine.py people.csv config.csv outfile.csv

For example,

python twine.py test_data/example1_people.csv test_data/example1_config.csv groups1.csv
diff groups1.csv test_data/example1_groups.csv

Potential Improvements

• Generalize fit score calculation to arbitrary number of people so that we can compute the best groups of arbitrary size

• Generalize attribute fit score calculation to allow for partial fits (i.e., attribute fit scores need not be only 0 or 1)

• Create test example with many more people to better check edge cases (e.g., the case where multiple pairs of people have identical fit scores so that we can ensure that ties are broken correctly)

Other Instructions

Find tests in test_twine.py. Run via python test_twine.py.

Also note that I have corrected the typos in the test data, which can be found in the test_data/ directory.