SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Merge remote-tracking branch 'huggingface/master'

Browse Source 
merge huggingface/master to update
This commit is contained in:
erenup
2019-08-17 12:05:57 +08:00
parent b219029c45 d8923270e6
commit e384ae2b9d
24 changed files with 1447 additions and 163 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
examples/run_glue.py
View File

@@ -13,7 +13,7 @@
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
""" Finetuning the library models for sequence classification on GLUE (Bert, XLM, XLNet)."""
""" Finetuning the library models for sequence classification on GLUE (Bert, XLM, XLNet, RoBERTa)."""
from __future__ import absolute_import, division, print_function
@@ -33,6 +33,9 @@ from tqdm import tqdm, trange
from pytorch_transformers import (WEIGHTS_NAME, BertConfig,
BertForSequenceClassification, BertTokenizer,
RobertaConfig,
RobertaForSequenceClassification,
RobertaTokenizer,
XLMConfig, XLMForSequenceClassification,
XLMTokenizer, XLNetConfig,
XLNetForSequenceClassification,
@@ -45,12 +48,13 @@ from utils_glue import (compute_metrics, convert_examples_to_features,
logger = logging.getLogger(__name__)
ALL_MODELS = sum((tuple(conf.pretrained_config_archive_map.keys()) for conf in (BertConfig, XLNetConfig, XLMConfig)), ())
ALL_MODELS = sum((tuple(conf.pretrained_config_archive_map.keys()) for conf in (BertConfig, XLNetConfig, XLMConfig, RobertaConfig)), ())
MODEL_CLASSES = {
'bert': (BertConfig, BertForSequenceClassification, BertTokenizer),
'xlnet': (XLNetConfig, XLNetForSequenceClassification, XLNetTokenizer),
'xlm': (XLMConfig, XLMForSequenceClassification, XLMTokenizer),
'roberta': (RobertaConfig, RobertaForSequenceClassification, RobertaTokenizer),
}
@@ -214,7 +218,7 @@ def evaluate(args, model, tokenizer, prefix=""):
with torch.no_grad():
inputs = {'input_ids': batch[0],
'attention_mask': batch[1],
'token_type_ids': batch[2] if args.model_type in ['bert', 'xlnet'] else None, # XLM don't use segment_ids
'token_type_ids': batch[2] if args.model_type in ['bert', 'xlnet'] else None, # XLM and RoBERTa don't use segment_ids
'labels': batch[3]}
outputs = model(**inputs)
tmp_eval_loss, logits = outputs[:2]
@@ -264,14 +268,20 @@ def load_and_cache_examples(args, task, tokenizer, evaluate=False):
else:
logger.info("Creating features from dataset file at %s", args.data_dir)
label_list = processor.get_labels()
if task in ['mnli', 'mnli-mm'] and args.model_type in ['roberta']:
# HACK(label indices are swapped in RoBERTa pretrained model)
label_list[1], label_list[2] = label_list[2], label_list[1]
examples = processor.get_dev_examples(args.data_dir) if evaluate else processor.get_train_examples(args.data_dir)
features = convert_examples_to_features(examples, label_list, args.max_seq_length, tokenizer, output_mode,
cls_token_at_end=bool(args.model_type in ['xlnet']), # xlnet has a cls token at the end
cls_token=tokenizer.cls_token,
sep_token=tokenizer.sep_token,
cls_token_segment_id=2 if args.model_type in ['xlnet'] else 0,
sep_token=tokenizer.sep_token,
sep_token_extra=bool(args.model_type in ['roberta']), # roberta uses an extra separator b/w pairs of sentences, cf. github.com/pytorch/fairseq/commit/1684e166e3da03f5b600dbb7855cb98ddfcd0805
pad_on_left=bool(args.model_type in ['xlnet']), # pad on the left for xlnet
pad_token_segment_id=4 if args.model_type in ['xlnet'] else 0)
pad_token=tokenizer.convert_tokens_to_ids([tokenizer.pad_token])[0],
pad_token_segment_id=4 if args.model_type in ['xlnet'] else 0,
)
if args.local_rank in [-1, 0]:
logger.info("Saving features into cached file %s", cached_features_file)
torch.save(features, cached_features_file)

29
examples/utils_glue.py
View File

@@ -390,10 +390,16 @@ class WnliProcessor(DataProcessor):
def convert_examples_to_features(examples, label_list, max_seq_length,
tokenizer, output_mode,
cls_token_at_end=False, pad_on_left=False,
cls_token='[CLS]', sep_token='[SEP]', pad_token=0,
sequence_a_segment_id=0, sequence_b_segment_id=1,
cls_token_segment_id=1, pad_token_segment_id=0,
cls_token_at_end=False,
cls_token='[CLS]',
cls_token_segment_id=1,
sep_token='[SEP]',
sep_token_extra=False,
pad_on_left=False,
pad_token=0,
pad_token_segment_id=0,
sequence_a_segment_id=0,
sequence_b_segment_id=1,
mask_padding_with_zero=True):
""" Loads a data file into a list of `InputBatch`s
`cls_token_at_end` define the location of the CLS token:
@@ -416,12 +422,14 @@ def convert_examples_to_features(examples, label_list, max_seq_length,
tokens_b = tokenizer.tokenize(example.text_b)
# Modifies `tokens_a` and `tokens_b` in place so that the total
# length is less than the specified length.
# Account for [CLS], [SEP], [SEP] with "- 3"
_truncate_seq_pair(tokens_a, tokens_b, max_seq_length - 3)
# Account for [CLS], [SEP], [SEP] with "- 3". " -4" for RoBERTa.
special_tokens_count = 4 if sep_token_extra else 3
_truncate_seq_pair(tokens_a, tokens_b, max_seq_length - special_tokens_count)
else:
# Account for [CLS] and [SEP] with "- 2"
if len(tokens_a) > max_seq_length - 2:
tokens_a = tokens_a[:(max_seq_length - 2)]
# Account for [CLS] and [SEP] with "- 2" and with "- 3" for RoBERTa.
special_tokens_count = 3 if sep_token_extra else 2
if len(tokens_a) > max_seq_length - special_tokens_count:
tokens_a = tokens_a[:(max_seq_length - special_tokens_count)]
# The convention in BERT is:
# (a) For sequence pairs:
@@ -442,6 +450,9 @@ def convert_examples_to_features(examples, label_list, max_seq_length,
# used as as the "sentence vector". Note that this only makes sense because
# the entire model is fine-tuned.
tokens = tokens_a + [sep_token]
if sep_token_extra:
# roberta uses an extra separator b/w pairs of sentences
tokens += [sep_token]
segment_ids = [sequence_a_segment_id] * len(tokens)
if tokens_b: