dnrops
/
transformers
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

adding s3 model tests with --runslow

This commit is contained in:
thomwolf 2019-04-17 11:58:27 +02:00
parent 8407429d74
commit 31d387604c
10 changed files with 106 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
.circleci/config.yml
View File

@ -9,7 +9,7 @@ jobs:
- run: sudo pip install --progress-bar off .
- run: sudo pip install pytest ftfy spacy
- run: sudo python -m spacy download en
- run: python -m pytest -sv tests/
- run: python -m pytest -sv tests/ --runslow
build_py2:
working_directory: ~/pytorch-pretrained-BERT
docker:
@ -20,7 +20,7 @@ jobs:
- run: sudo pip install pytest spacy
- run: sudo pip install ftfy==4.4.3
- run: sudo python -m spacy download en
- run: python -m pytest -sv tests/
- run: python -m pytest -sv tests/ --runslow
workflows:
version: 2
build_and_test:

19
tests/conftest.py Normal file
View File

@ -0,0 +1,19 @@
# content of conftest.py
import pytest
def pytest_addoption(parser):
parser.addoption(
"--runslow", action="store_true", default=False, help="run slow tests"
)
def pytest_collection_modifyitems(config, items):
if config.getoption("--runslow"):
# --runslow given in cli: do not skip slow tests
return
skip_slow = pytest.mark.skip(reason="need --runslow option to run")
for item in items:
if "slow" in item.keywords:
item.add_marker(skip_slow)

12
tests/modeling_gpt2_test.py
View File

@ -20,12 +20,14 @@ import os
import unittest
import json
import random
import shutil
import pytest
import torch
from pytorch_pretrained_bert import (GPT2Config, GPT2Model,
GPT2LMHeadModel, GPT2DoubleHeadsModel)
from pytorch_pretrained_bert.modeling_gpt2 import PRETRAINED_MODEL_ARCHIVE_MAP
class GPT2ModelTest(unittest.TestCase):
class GPT2ModelTester(object):
@ -185,6 +187,14 @@ class GPT2ModelTest(unittest.TestCase):
os.remove(json_file_path)
self.assertEqual(config_second.to_dict(), config_first.to_dict())
@pytest.mark.slow
def test_model_from_pretrained(self):
cache_dir = "/tmp/pytorch_pretrained_bert_test/"
for model_name in list(PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
model = GPT2Model.from_pretrained(model_name, cache_dir=cache_dir)
shutil.rmtree(cache_dir)
self.assertIsNotNone(model)
def run_tester(self, tester):
config_and_inputs = tester.prepare_config_and_inputs()
output_result = tester.create_gpt2_model(*config_and_inputs)

12
tests/modeling_openai_test.py
View File

@ -20,12 +20,14 @@ import os
import unittest
import json
import random
import shutil
import pytest
import torch
from pytorch_pretrained_bert import (OpenAIGPTConfig, OpenAIGPTModel,
OpenAIGPTLMHeadModel, OpenAIGPTDoubleHeadsModel)
from pytorch_pretrained_bert.modeling_openai import PRETRAINED_MODEL_ARCHIVE_MAP
class OpenAIGPTModelTest(unittest.TestCase):
class OpenAIGPTModelTester(object):
@ -197,6 +199,14 @@ class OpenAIGPTModelTest(unittest.TestCase):
os.remove(json_file_path)
self.assertEqual(config_second.to_dict(), config_first.to_dict())
@pytest.mark.slow
def test_model_from_pretrained(self):
cache_dir = "/tmp/pytorch_pretrained_bert_test/"
for model_name in list(PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
model = OpenAIGPTModel.from_pretrained(model_name, cache_dir=cache_dir)
shutil.rmtree(cache_dir)
self.assertIsNotNone(model)
def run_tester(self, tester):
config_and_inputs = tester.prepare_config_and_inputs()
output_result = tester.create_openai_model(*config_and_inputs)

11
tests/modeling_test.py
View File

@ -20,6 +20,8 @@ import os
import unittest
import json
import random
import shutil
import pytest
import torch
@ -27,6 +29,7 @@ from pytorch_pretrained_bert import (BertConfig, BertModel, BertForMaskedLM,
BertForNextSentencePrediction, BertForPreTraining,
BertForQuestionAnswering, BertForSequenceClassification,
BertForTokenClassification)
from pytorch_pretrained_bert.modeling import PRETRAINED_MODEL_ARCHIVE_MAP
class BertModelTest(unittest.TestCase):
@ -260,6 +263,14 @@ class BertModelTest(unittest.TestCase):
os.remove(json_file_path)
self.assertEqual(config_second.to_dict(), config_first.to_dict())
@pytest.mark.slow
def test_model_from_pretrained(self):
cache_dir = "/tmp/pytorch_pretrained_bert_test/"
for model_name in list(PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
model = BertModel.from_pretrained(model_name, cache_dir=cache_dir)
shutil.rmtree(cache_dir)
self.assertIsNotNone(model)
def run_tester(self, tester):
config_and_inputs = tester.prepare_config_and_inputs()
output_result = tester.create_bert_model(*config_and_inputs)

12
tests/modeling_transfo_xl_test.py
View File

@ -20,11 +20,13 @@ import os
import unittest
import json
import random
import shutil
import pytest
import torch
from pytorch_pretrained_bert import (TransfoXLConfig, TransfoXLModel, TransfoXLLMHeadModel)
from pytorch_pretrained_bert.modeling_transfo_xl import PRETRAINED_MODEL_ARCHIVE_MAP
class TransfoXLModelTest(unittest.TestCase):
class TransfoXLModelTester(object):
@ -195,6 +197,14 @@ class TransfoXLModelTest(unittest.TestCase):
os.remove(json_file_path)
self.assertEqual(config_second.to_dict(), config_first.to_dict())
@pytest.mark.slow
def test_model_from_pretrained(self):
cache_dir = "/tmp/pytorch_pretrained_bert_test/"
for model_name in list(PRETRAINED_MODEL_ARCHIVE_MAP.keys())[:1]:
model = TransfoXLModel.from_pretrained(model_name, cache_dir=cache_dir)
shutil.rmtree(cache_dir)
self.assertIsNotNone(model)
def run_tester(self, tester):
config_and_inputs = tester.prepare_config_and_inputs()

11
tests/tokenization_gpt2_test.py
View File

@ -17,8 +17,10 @@ from __future__ import absolute_import, division, print_function, unicode_litera
import os
import unittest
import json
import shutil
import pytest
from pytorch_pretrained_bert.tokenization_gpt2 import GPT2Tokenizer
from pytorch_pretrained_bert.tokenization_gpt2 import GPT2Tokenizer, PRETRAINED_VOCAB_ARCHIVE_MAP
class GPT2TokenizationTest(unittest.TestCase):
@ -64,6 +66,13 @@ class GPT2TokenizationTest(unittest.TestCase):
[tokenizer_2.encoder, tokenizer_2.decoder, tokenizer_2.bpe_ranks,
tokenizer_2.special_tokens, tokenizer_2.special_tokens_decoder])
@pytest.mark.slow
def test_tokenizer_from_pretrained(self):
cache_dir = "/tmp/pytorch_pretrained_bert_test/"
for model_name in list(PRETRAINED_VOCAB_ARCHIVE_MAP.keys())[:1]:
tokenizer = GPT2Tokenizer.from_pretrained(model_name, cache_dir=cache_dir)
shutil.rmtree(cache_dir)
self.assertIsNotNone(tokenizer)
if __name__ == '__main__':
unittest.main()

12
tests/tokenization_openai_test.py
View File

@ -17,8 +17,10 @@ from __future__ import absolute_import, division, print_function, unicode_litera
import os
import unittest
import json
import shutil
import pytest
from pytorch_pretrained_bert.tokenization_openai import OpenAIGPTTokenizer
from pytorch_pretrained_bert.tokenization_openai import OpenAIGPTTokenizer, PRETRAINED_VOCAB_ARCHIVE_MAP
class OpenAIGPTTokenizationTest(unittest.TestCase):
@ -64,6 +66,14 @@ class OpenAIGPTTokenizationTest(unittest.TestCase):
[tokenizer_2.encoder, tokenizer_2.decoder, tokenizer_2.bpe_ranks,
tokenizer_2.special_tokens, tokenizer_2.special_tokens_decoder])
@pytest.mark.slow
def test_tokenizer_from_pretrained(self):
cache_dir = "/tmp/pytorch_pretrained_bert_test/"
for model_name in list(PRETRAINED_VOCAB_ARCHIVE_MAP.keys())[:1]:
tokenizer = OpenAIGPTTokenizer.from_pretrained(model_name, cache_dir=cache_dir)
shutil.rmtree(cache_dir)
self.assertIsNotNone(tokenizer)
if __name__ == '__main__':
unittest.main()

11
tests/tokenization_test.py
View File

@ -17,12 +17,14 @@ from __future__ import absolute_import, division, print_function, unicode_litera
import os
import unittest
from io import open
import shutil
import pytest
from pytorch_pretrained_bert.tokenization import (BasicTokenizer,
BertTokenizer,
WordpieceTokenizer,
_is_control, _is_punctuation,
_is_whitespace)
_is_whitespace, PRETRAINED_VOCAB_ARCHIVE_MAP)
class TokenizationTest(unittest.TestCase):
@ -56,6 +58,13 @@ class TokenizationTest(unittest.TestCase):
self.assertListEqual(
tokenizer.convert_tokens_to_ids(tokens), [7, 4, 5, 10, 8, 9])
@pytest.mark.slow
def test_tokenizer_from_pretrained(self):
cache_dir = "/tmp/pytorch_pretrained_bert_test/"
for model_name in list(PRETRAINED_VOCAB_ARCHIVE_MAP.keys())[:1]:
tokenizer = BertTokenizer.from_pretrained(model_name, cache_dir=cache_dir)
shutil.rmtree(cache_dir)
self.assertIsNotNone(tokenizer)
def test_chinese(self):
tokenizer = BasicTokenizer()

11
tests/tokenization_transfo_xl_test.py
View File

@ -17,8 +17,10 @@ from __future__ import absolute_import, division, print_function, unicode_litera
import os
import unittest
from io import open
import shutil
import pytest
from pytorch_pretrained_bert.tokenization_transfo_xl import TransfoXLTokenizer
from pytorch_pretrained_bert.tokenization_transfo_xl import TransfoXLTokenizer, PRETRAINED_VOCAB_ARCHIVE_MAP
class TransfoXLTokenizationTest(unittest.TestCase):
@ -66,6 +68,13 @@ class TransfoXLTokenizationTest(unittest.TestCase):
tokenizer.tokenize(u" \tHeLLo ! how \n Are yoU ? "),
["HeLLo", "!", "how", "Are", "yoU", "?"])
@pytest.mark.slow
def test_tokenizer_from_pretrained(self):
cache_dir = "/tmp/pytorch_pretrained_bert_test/"
for model_name in list(PRETRAINED_VOCAB_ARCHIVE_MAP.keys())[:1]:
tokenizer = TransfoXLTokenizer.from_pretrained(model_name, cache_dir=cache_dir)
shutil.rmtree(cache_dir)
self.assertIsNotNone(tokenizer)
if __name__ == '__main__':
unittest.main()