File size: 10,203 Bytes
0240c6e |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 |
import os
import pickle
import numpy as np
from utils import CONFIG, RANDOM, load_datasets
from lxml import etree
from xml.etree.ElementTree import ElementTree
GRAM_TYPES = CONFIG['grammemes_types']
ROOT = CONFIG['publish_tests_path']
DICT_WORDS_PATH = CONFIG['dict_words_path']
NMB_DATA_PATH = CONFIG['numb_data_path']
with open(NMB_DATA_PATH, 'rb') as f:
numb_data = pickle.load(f)
with open(CONFIG['tags_path'], 'rb') as f:
tags = pickle.load(f)
is_lemma_dict = {tags[key]['i']: tags[key]['l'] for key in tags}
tag_index_order = {tags[tag]['i']: tags[tag]['o'] for tag in tags}
def release_gram_tests(items, key, cls_dic, result_path):
root = etree.Element('Tests')
for word in items:
y = np.argwhere(word['y'] == 1).ravel()
y = ';'.join([cls_dic[index] for index in y])
test = etree.Element("T")
test.set('x', word['src'])
test.set('y', y)
root.append(test)
rez_path = os.path.join(result_path, f'{key}.xml')
tree = ElementTree(root)
with open(rez_path, 'wb+') as f:
tree.write(f, xml_declaration=True, encoding='utf-8')
def release_main_tests(items, result_path, y_is_index=True):
root = etree.Element('Tests')
for word in items:
if y_is_index:
y = word['y']
else:
y = np.argwhere(word['y'] == 1).ravel()
y = ';'.join([str(index) for index in y])
test = etree.Element("T")
test.set('x', word['src'])
test.set('y', y)
root.append(test)
rez_path = os.path.join(result_path, f'main.xml')
tree = ElementTree(root)
with open(rez_path, 'wb+') as f:
tree.write(f, xml_declaration=True, encoding='utf-8')
def release_lemma_tests(items, result_path):
root = etree.Element('Tests')
for word in items:
test = etree.Element("T")
test.set('x', word['x_src'])
test.set('x_c', str(word['main_cls']))
test.set('y', word['y_src'])
root.append(test)
rez_path = os.path.join(result_path, 'lemma.xml')
tree = ElementTree(root)
with open(rez_path, 'wb+') as f:
tree.write(f, xml_declaration=True, encoding='utf-8')
def release_inflect_tests(items, result_path):
root = etree.Element('Tests')
for word in items:
test = etree.Element("T")
test.set('x', word['x_src'])
test.set('x_c', str(word['x_cls']))
test.set('y', word['y_src'])
test.set('y_c', str(word['y_cls']))
root.append(test)
rez_path = os.path.join(result_path, 'inflect.xml')
tree = ElementTree(root)
with open(rez_path, 'wb+') as f:
tree.write(f, xml_declaration=True, encoding='utf-8')
def release_nn_tests():
res_path = os.path.join(ROOT, 'Network')
for gram in GRAM_TYPES:
items = load_datasets(gram, 'test')
cls = GRAM_TYPES[gram]['classes']
dic = {cls[g_key]['index']: g_key for g_key in cls}
release_gram_tests(items, gram, dic, res_path)
release_main_tests(load_datasets('main', 'test'), res_path, False)
release_lemma_tests(load_datasets('lemma', 'test'), res_path)
release_inflect_tests(load_datasets('inflect', 'test'), res_path)
def merge_same_main(items):
rez_dict = {}
for item in items:
if item['src'] not in rez_dict:
rez_dict[item['src']] = []
rez_dict[item['src']].append(item)
rez_list = []
for text in rez_dict:
ys = [item['y'] for item in rez_dict[text]]
rez_list.append(dict(src=text, y=ys))
return rez_list
def release_dictionary_tests():
res_path = os.path.join(ROOT, 'Dict')
with open(DICT_WORDS_PATH, 'rb') as f:
words = pickle.load(f)
lexeme_dict = {}
for word in words:
if word['post'] == 'numb':
continue
if word['id'] not in lexeme_dict:
lexeme_dict[word['id']] = []
lexeme_dict[word['id']].append(word)
main = []
inflect = []
lemmas = []
for word_id in lexeme_dict:
lexeme_words = lexeme_dict[word_id]
for item in lexeme_words:
main.append(dict(src=item['text'], y=item['main']))
for word in lexeme_words:
lemma = word['lemma'] if 'lemma' in word else word['text']
if is_lemma_dict[word['main']]:
lemma = word['text']
lemmas.append(dict(
x_src=word['text'],
main_cls=word['main'],
y_src=lemma
))
un_cls_ids = []
rez_items = []
for item in sorted(lexeme_words, key=lambda x: x['index']):
if item['main'] in un_cls_ids:
continue
un_cls_ids.append(item['main'])
rez_items.append(item)
lexeme_words = rez_items
for i in range(0, len(lexeme_words) - 2):
main_word = lexeme_words[i]
for j in range(i, len(lexeme_words) - 1):
to_word = lexeme_words[j]
if to_word['main'] == main_word['main']:
continue
inflect.append(dict(
x_src=main_word['text'],
x_cls=main_word['main'],
y_src=to_word['text'],
y_cls=to_word['main'],
id=word_id
))
main = merge_same_main(main)
release_main_tests(main, res_path)
release_lemma_tests(lemmas, res_path)
release_inflect_tests(inflect, res_path)
def release_numb_tests():
res_path = os.path.join(ROOT, 'Numb')
main = []
inflect = []
lemmas = []
for val in numb_data['numbers']:
n_el = etree.Element("N")
n_el.set('v', str(val))
for tp in numb_data['numbers'][val]:
if tp == 'nar_end' or tp == 'lemma':
continue
items = numb_data['numbers'][val][tp]
lemma, _ = items[0]
for text, index in items:
main.append(dict(src=text, y=index))
lemmas.append(dict(
x_src=text,
main_cls=index,
y_src=lemma
))
un_cls_ids = []
rez_items = []
for item in items:
if item[1] in un_cls_ids:
continue
un_cls_ids.append(item[1])
rez_items.append(item)
items = rez_items
for i in range(0, len(items) - 2):
main_text, main_index = items[i]
for j in range(i, len(items) - 1):
to_text, to_index = items[j]
inflect.append(dict(
x_src=main_text,
x_cls=main_index,
y_src=to_text,
y_cls=to_index,
id=f"{val}{tp}"
))
main = merge_same_main(main)
release_main_tests(main, res_path)
release_lemma_tests(lemmas, res_path)
release_inflect_tests(inflect, res_path)
def release_nar_numb_tests():
res_path = os.path.join(ROOT, 'NarNumb')
main = []
inflect = []
lemmas = []
for val in numb_data['numbers']:
n_el = etree.Element("N")
n_el.set('v', str(val))
items = numb_data['numbers'][val]['nar_end']
lemma_id = numb_data['numbers'][val]['p'][0][1]
lemma = f"{val}-{items[lemma_id]}"
for index in items:
text = f"{val}-{items[index]}"
main.append(dict(src=text, y=index))
lemmas.append(dict(
x_src=text,
main_cls=index,
y_src=lemma
))
ids = list(items.keys())
for i in range(0, len(ids) - 2):
main_index = ids[i]
main_text = items[main_index]
main_text = f"{val}-{main_text}"
for j in range(i, len(items) - 1):
to_index = ids[j]
to_text = items[to_index]
to_text = f"{val}-{to_text}"
inflect.append(dict(
x_src=main_text,
x_cls=main_index,
y_src=to_text,
y_cls=to_index,
id=f"{val}nar"
))
main = merge_same_main(main)
release_main_tests(main, res_path)
release_lemma_tests(lemmas, res_path)
release_inflect_tests(inflect, res_path)
def release_reg_tests():
res_path = os.path.join(ROOT, 'Reg')
int_tag = None
romn_tag = None
unkn_tag = None
punct_tag = None
for tag in tags:
if 'int' in tag:
int_tag = tags[tag]['i']
elif 'romn' in tag:
romn_tag = tags[tag]['i']
elif 'unkn' in tag:
unkn_tag = tags[tag]['i']
elif 'punct' in tag:
punct_tag = tags[tag]['i']
main = []
puncts = ['.', ',', '?', '!', '_', '"', '(', ')', ':', ';', '-']
for p in puncts:
main.append(dict(src=p, y=punct_tag))
text = [p]
for _ in range(1, RANDOM.randint(2, 5)):
text.append(puncts[RANDOM.randint(0, len(puncts)-1)])
text = ''.join(text)
main.append(dict(src=text, y=punct_tag))
for i in range(100):
val = RANDOM.randint(0, 1000000)
main.append(dict(src=str(val), y=int_tag))
roms = ['i', 'iii', 'iv', 'iv', 'c', 'd', 'm', 'md', 'mi']
for rom in roms:
main.append(dict(src=rom, y=romn_tag))
main.append(dict(src=rom.upper(), y=romn_tag))
unkn = ['test', 'sdasdas', 'home']
for v in unkn:
main.append(dict(src=v, y=unkn_tag))
main.append(dict(src=v.upper(), y=unkn_tag))
lemmas = [dict(x_src=item['src'], main_cls=item['y'], y_src=item['src']) for item in main]
inflect = [dict(x_src=item['src'], x_cls=item['y'], y_src=item['src'], y_cls=item['y'], id=item['src']) for item in main]
main = merge_same_main(main)
release_main_tests(main, res_path)
release_lemma_tests(lemmas, res_path)
release_inflect_tests(inflect, res_path)
release_reg_tests()
release_nar_numb_tests()
release_numb_tests()
release_dictionary_tests()
release_nn_tests()
|