def test_not_in_wordnet(self): self.assertEqual(not_in_wordnet("hot"), False)
# r = r.encode('ascii', 'ignore').decode('ascii') # bytes object is not being handled by most_similar function try: n = 0 sims = model.most_similar(positive=[r], topn=200) for s in sims: n = n + 1 print("counter: " + str(counter) + "-" + str(n)) # s = (s[0].encode("ascii", 'ignore'), s[1]) hit = False search = True if same_stem(s[0], r): search = False printout(",".join(['same stem', s[0], r, str(s[1]), str(n)])) if not_in_wordnet(r): search = False printout(",".join( ['not_in_wordnet', s[0], r, str(s[1]), str(n)])) if search: hyper = hypernomous(s[0], r) hypo = hyoponomous(s[0], r) syno = synononymous(s[0], r) holo = holonymous(s[0], r) mero = meronymous(s[0], r) if hyper > 0 and hyper < 1: printout(",".join( ['hyper', s[0], r, str(s[1]), str(n),
for r in words: counter = counter + 1 r = r.encode('ascii', 'ignore') try: k = 0 sims = model.most_similar(positive=[r], topn=200) for s in sims: k = k + 1 for pos in parts_of_speech: s = (s[0].encode("ascii", 'ignore'), s[1]) hit = False search = True if same_stem(s[0], r): search = False printout(",".join(['same stem', pos, str(model.similarity(s[0], r)), s[0], r, str(s[1]), str(counter) + "-" + str(k), str(1)])) if not_in_wordnet(r, pos): search = False printout(",".join(['not_in_wordnet', pos, str(model.similarity(s[0], r)), s[0], r, str(s[1]), str(counter) + "-" + str(k), str(1)])) if search: hyper = hypernomous(r, s[0], pos) hypo = hyoponomous(r, s[0], pos) syno = synononymous(r, s[0], pos) holo = holonymous(r, s[0], pos) mero = meronymous(r, s[0], pos) all_possible_hyper = get_all_possible_hypernyms(r, s[0], pos) all_possible_hypo = get_all_possible_hyponyms(r, s[0], pos) all_possible_mero = get_all_possible_meronyms(r, s[0], pos) all_possible_holo = get_all_possible_holonyms(r, s[0], pos) all_possible_syns = len(wn.synsets(r, pos=pos)) * len(wn.synsets(s[0], pos=pos)) tracker = str(counter) + "-" + str(k) printout(",".join(['hyper', pos, str(model.similarity(s[0], r)), r.replace(',', ""), s[0].replace(',', ""), str(s[1]), tracker, str(hyper), str(all_possible_hyper)]))
for s in sims: k = k + 1 for pos in parts_of_speech: s = (s[0].encode("ascii", 'ignore'), s[1]) hit = False search = True if same_stem(s[0], r): search = False printout(",".join([ 'same stem', pos, str(model.similarity(s[0], r)), s[0], r, str(s[1]), str(counter) + "-" + str(k), str(1) ])) if not_in_wordnet(r, pos): search = False printout(",".join([ 'not_in_wordnet', pos, str(model.similarity(s[0], r)), s[0], r, str(s[1]), str(counter) + "-" + str(k), str(1) ])) if search: hyper = hypernomous(r, s[0], pos) hypo = hyoponomous(r, s[0], pos) syno = synononymous(r, s[0], pos) holo = holonymous(r, s[0], pos) mero = meronymous(r, s[0], pos) all_possible_hyper = get_all_possible_hypernyms(
for r in words: counter = counter + 1 r = r.encode('ascii', 'ignore') try: n = 0 sims = model.most_similar(positive=[r], topn=200) for s in sims: n = n + 1 print "counter: " + str(counter) + "-" + str(n) s = (s[0].encode("ascii", 'ignore'), s[1]) hit = False search = True if same_stem(s[0], r): search = False printout(",".join(['same stem', s[0], r, str(s[1]), str(n)])) if not_in_wordnet(r): search = False printout(",".join(['not_in_wordnet', s[0], r, str(s[1]), str(n)])) if search: hyper = hypernomous(s[0], r) hypo = hyoponomous(s[0], r) syno = synononymous(s[0], r) holo = holonymous(s[0], r) mero = meronymous(s[0], r) if hyper > 0 and hyper < 1: printout(",".join(['hyper', s[0], r, str(s[1]), str(n), str(hyper)])) if hypo > 0 and hypo < 1: printout(",".join(['hypo', s[0], r, str(s[1]), str(n), str(hypo)])) if syno > 0 and syno < 1: printout(",".join(['syn', s[0], r, str(s[1]), str(n), str(syno)])) if holo > 0 and holo < 1: