row_count = 0;
try:
for row in reader:
row_count = row_count + 1;
item_num = int(len(row) / 8)
for i in range(1, item_num + 1):
start_sub = 8 * (i - 1)
if row[start_sub + 1] == '':
continue
map_name = row[start_sub+1];
original_alignment = row[start_sub + 2];
dict_alignment = row[start_sub + 3];
score = classify_gbc.sequence2(original_alignment, dict_alignment)
if score == sys.maxsize:
# print('Invalid alignment.')
pass
else:
writer.writerow({'map_name': map_name, 'org_text': original_alignment.replace(',', ''),
'dict_text': dict_alignment.replace('-', '').replace(',', ''),
'score': score, 'x': row[start_sub + 4], 'y': row[start_sub + 5],
'w': row[start_sub + 6], 'h': row[start_sub + 7]})
# pass
# print('score: ', score)
print('finish row ', row_count)
except UnicodeDecodeError:
print(row)
import classify_gbc
import io
if __name__ == '__main__':
with io.open('python_input.csv', 'r', newline='', encoding="mbcs") as fp:
reader = csv.reader(fp)
count = 0
for row in reader:
print(count)
count += 1
row_size = len(row)
iter = row_size / 8
offset = 0
for j in range(0, int(iter)):
offset = j
map_id = 0 if row[1 + 8 * offset] == '' else int(row[1 + 8 * offset])
truth = row[2 + 8 * offset].lower()
dict = row[3 + 8 * offset].lower()
x = 0 if row[4 + 8 * offset] == '' else int(row[4 + 8 * offset])
y = 0 if row[5 + 8 * offset] == '' else int(row[5 + 8 * offset])
w = 0 if row[6 + 8 * offset] == '' else int(row[6 + 8 * offset])
h = 0 if row[7 + 8 * offset] == '' else int(row[7 + 8 * offset])
if len(truth) == 0 or len(dict) == 0:
continue
classify_gbc.sequence2(truth, dict, x, y, w, h, map_id)
lines = cf.readlines()
count = 0
candidates = []
word_score = {}
# Compute the score for all the candidates.
while count + 5 < len(lines):
# print('Computing score for candiate: %d' % int(count / 6))
original_phrase = lines[count + 1]
original_word = lines[count + 2][:-1].replace('-', '')
alignment0 = lines[count + 2][:-1]
alignment1 = lines[count + 3][:-1]
idx = int(lines[count + 4][:-1])
score = classify_gbc.sequence2(alignment0, alignment1)
if original_phrase not in word_score:
word_score[original_phrase] = {}
if original_word not in word_score[original_phrase]:
word_score[original_phrase][original_word] = {}
word_score[original_phrase][original_word]['max'] = -10000
word_score[original_phrase][original_word]['cand'] = []
dict_text = alignment1.replace('-', '')
word_score[original_phrase][original_word]['cand'].append((idx, dict_text, score))
if score > word_score[original_phrase][original_word]['max']:
word_score[original_phrase][original_word]['max'] = score
count += 6
import classify_gbc
if __name__ == '__main__':
score = classify_gbc.sequence2('hill', 'hill')
print(score)
