# Check if the given column values are normal!
if len(columns) > col_key and len(columns) > col_value:
the_key = columns[col_key].strip()
the_value = columns[col_value].strip()
# Add the new pair in the dict. If the key is already
# existing, it will be updated with the new value!
my_dict[the_key] = the_value
else:
print("Warning. Bad number of columns for line: {0}".format(line))
return my_dict
# ----------------------------------------------------------------------------
if __name__ == '__main__':
lines = read_file(my_file)
# before doing something, check the data!
if not len(lines):
print('Hum... the file was empty!')
sys.exit(0)
sampa_dict = extract_dict_from_lines(lines, 1, 2)
my_list = ['a', 'b', 'c', 'd', 'e', 'f', 'E', 'g', 'a~', 'S']
for phone in my_list:
if phone in sampa_dict:
print("Sampa phoneme {:s} is IPA {:s}.".format(phone, sampa_dict[phone]))
else:
print("Sampa phoneme {:s} has no IPA!".format(phone))
for d in documents:
if item in d:
dw += 1.0
if dw == 0.0:
return 0.0
return tf * (math.log(D / dw))
# ---------------------------------------------------------------------------
# Main program
# ---------------------------------------------------------------------------
if __name__ == '__main__':
phones1 = read_file(corpus1)
phones2 = read_file(corpus2)
counter1 = collections.Counter(phones1)
counter2 = collections.Counter(phones2)
# Hapax
hapax1 = [k for k in counter1.keys() if counter1[k] == 1]
hapax2 = [k for k in counter2.keys() if counter2[k] == 1]
print("Corpus 1, Number of hapax: {:d}.".format(len(hapax1)))
print("Corpus 2, Number of hapax: {:d}.".format(len(hapax2)))
# Zipf law
ranks1 = get_ranks(counter1)
ranks2 = get_ranks(counter2)
for t in ['@', 'e', "E"]: