yurakuratov commited on
Commit
4633e5a
·
verified ·
1 Parent(s): 0dfffad

update model imports to be compatible with transformers 4.56.0

Browse files
Files changed (1) hide show
  1. modeling_bert.py +8 -8
modeling_bert.py CHANGED
@@ -48,12 +48,8 @@ from transformers.modeling_outputs import (
48
  SequenceClassifierOutput,
49
  TokenClassifierOutput,
50
  )
51
- from transformers.modeling_utils import (
52
- PreTrainedModel,
53
- apply_chunking_to_forward,
54
- find_pruneable_heads_and_indices,
55
- prune_linear_layer,
56
- )
57
  from transformers.utils import logging
58
  from transformers.models.bert.configuration_bert import BertConfig
59
 
@@ -1843,6 +1839,9 @@ class BertForSequenceClassification(BertPreTrainedModel):
1843
 
1844
  loss = None
1845
  if labels is not None:
 
 
 
1846
  if self.config.problem_type is None:
1847
  if self.num_labels == 1:
1848
  self.config.problem_type = "regression"
@@ -1850,7 +1849,7 @@ class BertForSequenceClassification(BertPreTrainedModel):
1850
  self.config.problem_type = "single_label_classification"
1851
  else:
1852
  self.config.problem_type = "multi_label_classification"
1853
-
1854
  if self.config.problem_type == "regression":
1855
  loss_fct = MSELoss()
1856
  if self.num_labels == 1:
@@ -1858,6 +1857,8 @@ class BertForSequenceClassification(BertPreTrainedModel):
1858
  else:
1859
  loss = loss_fct(logits, labels)
1860
  elif self.config.problem_type == "single_label_classification":
 
 
1861
  loss_fct = CrossEntropyLoss()
1862
  loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
1863
  elif self.config.problem_type == "multi_label_classification":
@@ -1987,7 +1988,6 @@ class BertForTokenClassification(BertPreTrainedModel):
1987
  self.config = config
1988
  if getattr(self.config, 'problem_type', None) is None:
1989
  self.config.problem_type = 'single_label_classification'
1990
-
1991
  self.bert = BertModel(config, add_pooling_layer=False)
1992
  classifier_dropout = (
1993
  config.classifier_dropout if config.classifier_dropout is not None else config.hidden_dropout_prob
 
48
  SequenceClassifierOutput,
49
  TokenClassifierOutput,
50
  )
51
+ from transformers.modeling_utils import PreTrainedModel
52
+ from transformers.pytorch_utils import apply_chunking_to_forward, find_pruneable_heads_and_indices, prune_linear_layer
 
 
 
 
53
  from transformers.utils import logging
54
  from transformers.models.bert.configuration_bert import BertConfig
55
 
 
1839
 
1840
  loss = None
1841
  if labels is not None:
1842
+ # print (f"self.config.problem_type from init: {self.config.problem_type}")
1843
+ # print (f"self.num_labels from init: {self.num_labels}")
1844
+ # print (f"labels.dtype {labels.dtype}")
1845
  if self.config.problem_type is None:
1846
  if self.num_labels == 1:
1847
  self.config.problem_type = "regression"
 
1849
  self.config.problem_type = "single_label_classification"
1850
  else:
1851
  self.config.problem_type = "multi_label_classification"
1852
+ # print (f"self.config.problem_type from init: {self.config.problem_type}")
1853
  if self.config.problem_type == "regression":
1854
  loss_fct = MSELoss()
1855
  if self.num_labels == 1:
 
1857
  else:
1858
  loss = loss_fct(logits, labels)
1859
  elif self.config.problem_type == "single_label_classification":
1860
+ # print (logits)
1861
+ # print (labels)
1862
  loss_fct = CrossEntropyLoss()
1863
  loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
1864
  elif self.config.problem_type == "multi_label_classification":
 
1988
  self.config = config
1989
  if getattr(self.config, 'problem_type', None) is None:
1990
  self.config.problem_type = 'single_label_classification'
 
1991
  self.bert = BertModel(config, add_pooling_layer=False)
1992
  classifier_dropout = (
1993
  config.classifier_dropout if config.classifier_dropout is not None else config.hidden_dropout_prob