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

Making TF OpenAI GPT model compliant with AMP and XLA (#10261)

Browse Source 
* Fix AMP and XLA

* Remove useless var
This commit is contained in:
Julien Plu
2021-02-19 15:33:25 +01:00
committed by GitHub
parent 3e116ed331
commit 34df26ec3a
2 changed files with 29 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
tests/test_modeling_tf_openai.py
View File

@@ -246,14 +246,6 @@ class TFOpenAIGPTModelTest(TFModelTesterMixin, unittest.TestCase):
config_and_inputs = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_openai_gpt_for_sequence_classification(*config_and_inputs)
def test_mixed_precision(self):
# TODO JP: Make OpenAIGPT float16 compliant
pass
def test_xla_mode(self):
# TODO JP: Make OpenAIGPT XLA compliant
pass
@slow
def test_model_from_pretrained(self):
for model_name in TF_OPENAI_GPT_PRETRAINED_MODEL_ARCHIVE_LIST[:1]: