Fast transformers import part 1 (#9441)

* Don't import libs to check they are available

* Don't import integrations at init

* Add importlib_metdata to deps

* Remove old vars references

* Avoid syntax error

* Adapt testing utils

* Try to appease torchhub

* Add dependency

* Remove more private variables

* Fix typo

* Another typo

* Refine the tf availability test

tests/test_tokenization_mbart.py

@@ -16,17 +16,13 @@ import tempfile
 import unittest
 
 from transformers import SPIECE_UNDERLINE, BatchEncoding, MBartTokenizer, MBartTokenizerFast, is_torch_available
-from transformers.testing_utils import (
-    _sentencepiece_available,
-    require_sentencepiece,
-    require_tokenizers,
-    require_torch,
-)
+from transformers.file_utils import is_sentencepiece_available
+from transformers.testing_utils import require_sentencepiece, require_tokenizers, require_torch
 
 from .test_tokenization_common import TokenizerTesterMixin
 
 
-if _sentencepiece_available:
+if is_sentencepiece_available():
     from .test_tokenization_xlm_roberta import SAMPLE_VOCAB