script for finding functional dependencies in MorphGNT columns

dep.py

#!/usr/bin/env python3

import argparse
import collections
import glob

parser = argparse.ArgumentParser(description="count (and optionally list) the entries where the determinant columns do not functionally determine the dependent columns.")
parser.add_argument("-v", "--verbose", help="output full results", action="store_true")
parser.add_argument("determinant", help="comma-separated list of columns")
parser.add_argument("dependent", help="comma-separated list of columns")
args = parser.parse_args()

values = collections.defaultdict(set)
for filename in glob.glob("*.txt"):
    with open(filename) as f:
        for line in f:
            cells = line.strip().split()
            left_cell = " ".join(cells[int(col) - 1] for col in args.determinant.split(","))
            right_cell = " ".join(cells[int(col) - 1] for col in args.dependent.split(","))
            values[left_cell].add(right_cell)

count = 0
for k, v in values.items():
    if len(v) > 1:
        count += 1
        if args.verbose:
            print(k, v)
print(count)