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

Fix cached_file in offline mode for cached non-existing files (#19206)

Browse Source 
* Fix cached_file in offline mode for cached non-existing files

* Add tests

* Test with offline mode
This commit is contained in:
Sylvain Gugger
2022-09-26 18:01:00 -04:00
committed by GitHub
parent ca0886395b
commit a32f97c37d
4 changed files with 114 additions and 50 deletions
Download Patch File Download Diff File Expand all files Collapse all files

48
tests/utils/test_file_utils.py
View File

@@ -15,28 +15,14 @@
import contextlib
import importlib
import io
import json
import tempfile
import unittest
from pathlib import Path
import transformers
# Try to import everything from transformers to ensure every object can be loaded.
from transformers import * # noqa F406
from transformers.testing_utils import DUMMY_UNKNOWN_IDENTIFIER
from transformers.utils import (
FLAX_WEIGHTS_NAME,
TF2_WEIGHTS_NAME,
WEIGHTS_NAME,
ContextManagers,
find_labels,
get_file_from_repo,
has_file,
is_flax_available,
is_tf_available,
is_torch_available,
)
from transformers.utils import ContextManagers, find_labels, is_flax_available, is_tf_available, is_torch_available
MODEL_ID = DUMMY_UNKNOWN_IDENTIFIER
@@ -77,38 +63,6 @@ class TestImportMechanisms(unittest.TestCase):
assert importlib.util.find_spec("transformers") is not None
class GetFromCacheTests(unittest.TestCase):
def test_has_file(self):
self.assertTrue(has_file("hf-internal-testing/tiny-bert-pt-only", WEIGHTS_NAME))
self.assertFalse(has_file("hf-internal-testing/tiny-bert-pt-only", TF2_WEIGHTS_NAME))
self.assertFalse(has_file("hf-internal-testing/tiny-bert-pt-only", FLAX_WEIGHTS_NAME))
def test_get_file_from_repo_distant(self):
# `get_file_from_repo` returns None if the file does not exist
self.assertIsNone(get_file_from_repo("bert-base-cased", "ahah.txt"))
# The function raises if the repository does not exist.
with self.assertRaisesRegex(EnvironmentError, "is not a valid model identifier"):
get_file_from_repo("bert-base-case", "config.json")
# The function raises if the revision does not exist.
with self.assertRaisesRegex(EnvironmentError, "is not a valid git identifier"):
get_file_from_repo("bert-base-cased", "config.json", revision="ahaha")
resolved_file = get_file_from_repo("bert-base-cased", "config.json")
# The name is the cached name which is not very easy to test, so instead we load the content.
config = json.loads(open(resolved_file, "r").read())
self.assertEqual(config["hidden_size"], 768)
def test_get_file_from_repo_local(self):
with tempfile.TemporaryDirectory() as tmp_dir:
filename = Path(tmp_dir) / "a.txt"
filename.touch()
self.assertEqual(get_file_from_repo(tmp_dir, "a.txt"), str(filename))
self.assertIsNone(get_file_from_repo(tmp_dir, "b.txt"))
class GenericUtilTests(unittest.TestCase):
@unittest.mock.patch("sys.stdout", new_callable=io.StringIO)
def test_context_managers_no_context(self, mock_stdout):