diff --git a/metrics/metrics.py b/metrics/metrics.py index 81e2c6985ec50eaef3fef2bef438573938031438..521e0fdb4ac6776ced971c7e83f9596eaf8f59d1 100644 --- a/metrics/metrics.py +++ b/metrics/metrics.py @@ -356,30 +356,3 @@ if __name__ == "__main__": for line in csv.reader(sys.stdin): name, xsampa, orthography = line writer.writerow(evaluate(name, xsampa)) - for filename in [name + "_m.csv", name + "_f.csv"]: - with open("webmaus/" + filename, "w") as webmaus_input: - orthography_words = orthography.split(" ") - xsampa_words = xsampa.split(" ") - for orthography_word, xsampa_word in zip( - orthography_words, xsampa_words - ): - preprocessed = re.sub( - "(.):", - r"\1\1", # a: → aa, d: → dd etc. - xsampa_word.replace( - ".", "" - ) # we don't care about syllable breaks - .replace('"', "") # we don't care about primary stress - .replace("%", ""), # we don't care about secondary stress - ) - tokens = list( - itertools.chain( - *(tokenize(word) for word in preprocessed.split()) - ) - ) - webmaus_input.write( - orthography_word - + ";" - + " ".join(tokens).replace("qX", "q X") - + "\n" - )