File size: 9,300 Bytes
b410583 |
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 |
import json
def add_lang_by_task(target_str, task, sub_task):
if task == 'summarize':
target_str = '<en> ' + target_str
elif task == 'refine':
target_str = '<java> ' + target_str
elif task == 'translate':
if sub_task == 'java-cs':
target_str = '<c_sharp> ' + target_str
else:
target_str = '<java> ' + target_str
elif task == 'concode':
target_str = '<java> ' + target_str
elif task == 'defect':
target_str = target_str
return target_str
def convert_examples_to_features(item):
example, example_index, tokenizer, args, stage = item
if args.model_type in ['t5', 'codet5'] and args.add_task_prefix:
if args.sub_task != 'none':
source_str = "{} {}: {}".format(args.task, args.sub_task, example.source)
else:
source_str = "{}: {}".format(args.task, example.source)
else:
source_str = example.source
source_str = source_str.replace('</s>', '<unk>')
source_ids = tokenizer.encode(source_str, max_length=args.max_source_length, padding='max_length', truncation=True)
assert source_ids.count(tokenizer.eos_token_id) == 1
if stage == 'test':
target_ids = []
else:
target_str = example.target
if args.add_lang_ids:
target_str = add_lang_by_task(example.target, args.task, args.sub_task)
if args.task in ['defect', 'clone']:
if target_str == 0:
target_str = 'false'
elif target_str == 1:
target_str = 'true'
else:
raise NameError
target_str = target_str.replace('</s>', '<unk>')
target_ids = tokenizer.encode(target_str, max_length=args.max_target_length, padding='max_length',
truncation=True)
assert target_ids.count(tokenizer.eos_token_id) == 1
return InputFeatures(
example_index,
source_ids,
target_ids,
url=example.url
)
def convert_clone_examples_to_features(item):
example, example_index, tokenizer, args = item
if args.model_type in ['t5', 'codet5'] and args.add_task_prefix:
source_str = "{}: {}".format(args.task, example.source)
target_str = "{}: {}".format(args.task, example.target)
else:
source_str = example.source
target_str = example.target
code1 = tokenizer.encode(source_str, max_length=args.max_source_length, padding='max_length', truncation=True)
code2 = tokenizer.encode(target_str, max_length=args.max_source_length, padding='max_length', truncation=True)
source_ids = code1 + code2
return CloneInputFeatures(example_index, source_ids, example.label, example.url1, example.url2)
def convert_defect_examples_to_features(item):
example, example_index, tokenizer, args = item
if args.model_type in ['t5', 'codet5'] and args.add_task_prefix:
source_str = "{}: {}".format(args.task, example.source)
else:
source_str = example.source
code = tokenizer.encode(source_str, max_length=args.max_source_length, padding='max_length', truncation=True)
return DefectInputFeatures(example_index, code, example.target)
class CloneInputFeatures(object):
"""A single training/test features for a example."""
def __init__(self,
example_id,
source_ids,
label,
url1,
url2
):
self.example_id = example_id
self.source_ids = source_ids
self.label = label
self.url1 = url1
self.url2 = url2
class DefectInputFeatures(object):
"""A single training/test features for a example."""
def __init__(self,
example_id,
source_ids,
label
):
self.example_id = example_id
self.source_ids = source_ids
self.label = label
class InputFeatures(object):
"""A single training/test features for a example."""
def __init__(self,
example_id,
source_ids,
target_ids,
url=None
):
self.example_id = example_id
self.source_ids = source_ids
self.target_ids = target_ids
self.url = url
class Example(object):
"""A single training/test example."""
def __init__(self,
idx,
source,
target,
url=None,
task='',
sub_task=''
):
self.idx = idx
self.source = source
self.target = target
self.url = url
self.task = task
self.sub_task = sub_task
class CloneExample(object):
"""A single training/test example."""
def __init__(self,
code1,
code2,
label,
url1,
url2
):
self.source = code1
self.target = code2
self.label = label
self.url1 = url1
self.url2 = url2
def read_translate_examples(filename, data_num):
"""Read examples from filename."""
examples = []
assert len(filename.split(',')) == 2
src_filename = filename.split(',')[0]
trg_filename = filename.split(',')[1]
idx = 0
with open(src_filename) as f1, open(trg_filename) as f2:
for line1, line2 in zip(f1, f2):
src = line1.strip()
trg = line2.strip()
examples.append(
Example(
idx=idx,
source=src,
target=trg,
)
)
idx += 1
if idx == data_num:
break
return examples
def read_refine_examples(filename, data_num):
"""Read examples from filename."""
examples = []
assert len(filename.split(',')) == 2
src_filename = filename.split(',')[0]
trg_filename = filename.split(',')[1]
idx = 0
with open(src_filename) as f1, open(trg_filename) as f2:
for line1, line2 in zip(f1, f2):
examples.append(
Example(
idx=idx,
source=line1.strip(),
target=line2.strip(),
)
)
idx += 1
if idx == data_num:
break
return examples
def read_concode_examples(filename, data_num):
"""Read examples from filename."""
examples = []
with open(filename) as f:
for idx, line in enumerate(f):
x = json.loads(line)
examples.append(
Example(
idx=idx,
source=x["nl"].strip(),
target=x["code"].strip()
)
)
idx += 1
if idx == data_num:
break
return examples
def read_summarize_examples(filename, data_num):
"""Read examples from filename."""
examples = []
with open(filename, encoding="utf-8") as f:
for idx, line in enumerate(f):
line = line.strip()
js = json.loads(line)
if 'idx' not in js:
js['idx'] = idx
code = ' '.join(js['code_tokens']).replace('\n', ' ')
code = ' '.join(code.strip().split())
nl = ' '.join(js['docstring_tokens']).replace('\n', '')
nl = ' '.join(nl.strip().split())
examples.append(
Example(
idx=idx,
source=code,
target=nl,
)
)
if idx + 1 == data_num:
break
return examples
def read_defect_examples(filename, data_num):
"""Read examples from filename."""
examples = []
with open(filename, encoding="utf-8") as f:
for idx, line in enumerate(f):
line = line.strip()
js = json.loads(line)
code = ' '.join(js['func'].split())
examples.append(
Example(
idx=js['idx'],
source=code,
target=js['target']
)
)
if idx + 1 == data_num:
break
return examples
def read_clone_examples(filename, data_num):
"""Read examples from filename."""
index_filename = filename
url_to_code = {}
with open('/'.join(index_filename.split('/')[:-1]) + '/data.jsonl') as f:
for line in f:
line = line.strip()
js = json.loads(line)
code = ' '.join(js['func'].split())
url_to_code[js['idx']] = code
data = []
with open(index_filename) as f:
idx = 0
for line in f:
line = line.strip()
url1, url2, label = line.split('\t')
if url1 not in url_to_code or url2 not in url_to_code:
continue
if label == '0':
label = 0
else:
label = 1
data.append(CloneExample(url_to_code[url1], url_to_code[url2], label, url1, url2))
idx += 1
if idx == data_num:
break
return data
|