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

[ci] fix 3 remaining slow GPU failures (#4584)

Browse Source
This commit is contained in:
Sam Shleifer
2020-05-25 19:20:50 -04:00
committed by GitHub
parent 365d452d4d
commit b86e42e0ac
4 changed files with 5 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
src/transformers/configuration_distilbert.py
View File

@@ -73,10 +73,10 @@ class DistilBertConfig(PretrainedConfig):
The standard deviation of the truncated_normal_initializer for initializing all weight matrices. The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
qa_dropout (:obj:`float`, optional, defaults to 0.1): qa_dropout (:obj:`float`, optional, defaults to 0.1):
The dropout probabilities used in the question answering model The dropout probabilities used in the question answering model
:class:`~tranformers.DistilBertForQuestionAnswering`. :class:`~transformers.DistilBertForQuestionAnswering`.
seq_classif_dropout (:obj:`float`, optional, defaults to 0.2): seq_classif_dropout (:obj:`float`, optional, defaults to 0.2):
The dropout probabilities used in the sequence classification model The dropout probabilities used in the sequence classification model
:class:`~tranformers.DistilBertForSequenceClassification`. :class:`~transformers.DistilBertForSequenceClassification`.
Example:: Example::

2
src/transformers/modeling_encoder_decoder.py
View File

@@ -125,7 +125,7 @@ class EncoderDecoderModel(PreTrainedModel):
Examples:: Examples::
from tranformers import EncoderDecoder from transformers import EncoderDecoder
model = EncoderDecoder.from_encoder_decoder_pretrained('bert-base-uncased', 'bert-base-uncased') # initialize Bert2Bert model = EncoderDecoder.from_encoder_decoder_pretrained('bert-base-uncased', 'bert-base-uncased') # initialize Bert2Bert
""" """

2
tests/test_modeling_bart.py
View File

@@ -240,7 +240,7 @@ class BartTranslationTests(unittest.TestCase):
with torch.no_grad(): with torch.no_grad():
logits, *other_stuff = model(**self.net_input) logits, *other_stuff = model(**self.net_input)
expected_slice = torch.tensor([9.0078, 10.1113, 14.4787]) expected_slice = torch.tensor([9.0078, 10.1113, 14.4787], device=torch_device)
result_slice = logits[0][0][:3] result_slice = logits[0][0][:3]
self.assertTrue(torch.allclose(expected_slice, result_slice, atol=TOLERANCE)) self.assertTrue(torch.allclose(expected_slice, result_slice, atol=TOLERANCE))

2
tests/test_modeling_tf_electra.py
View File

@@ -222,6 +222,6 @@ class TFElectraModelTest(TFModelTesterMixin, unittest.TestCase):
@slow @slow
def test_model_from_pretrained(self): def test_model_from_pretrained(self):
# for model_name in list(TF_ELECTRA_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]: # for model_name in list(TF_ELECTRA_PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
for model_name in ["electra-small-discriminator"]: for model_name in ["google/electra-small-discriminator"]:
model = TFElectraModel.from_pretrained(model_name) model = TFElectraModel.from_pretrained(model_name)
self.assertIsNotNone(model) self.assertIsNotNone(model)