[Test refactor 1/5] Per-folder tests reorganization (#15725)

* Per-folder tests reorganization
Co-authored-by: sgugger <sylvain.gugger@gmail.com>
Co-authored-by: Stas Bekman <stas@stason.org>

tests/test_modeling_flax_gpt_neo.py → tests/gpt_neo/test_modeling_flax_gpt_neo.py

@@ -22,8 +22,8 @@ import transformers
 from transformers import GPT2Tokenizer, GPTNeoConfig, is_flax_available, is_torch_available
 from transformers.testing_utils import is_pt_flax_cross_test, require_flax, slow
 
-from .test_generation_flax_utils import FlaxGenerationTesterMixin
-from .test_modeling_flax_common import FlaxModelTesterMixin, ids_tensor, random_attention_mask
+from ..generation.test_generation_flax_utils import FlaxGenerationTesterMixin
+from ..test_modeling_flax_common import FlaxModelTesterMixin, ids_tensor, random_attention_mask
 
 
 if is_flax_available():

tests/test_modeling_gpt_neo.py → tests/gpt_neo/test_modeling_gpt_neo.py

@@ -21,9 +21,9 @@ from transformers import GPTNeoConfig, is_torch_available
 from transformers.file_utils import cached_property
 from transformers.testing_utils import require_torch, slow, torch_device
 
-from .test_configuration_common import ConfigTester
-from .test_generation_utils import GenerationTesterMixin
-from .test_modeling_common import ModelTesterMixin, floats_tensor, ids_tensor, random_attention_mask
+from ..generation.test_generation_utils import GenerationTesterMixin
+from ..test_configuration_common import ConfigTester
+from ..test_modeling_common import ModelTesterMixin, floats_tensor, ids_tensor, random_attention_mask
 
 
 if is_torch_available():
@@ -95,7 +95,7 @@ class GPTNeoModelTester:
         self.attention_types = attention_types
 
     def get_large_model_config(self):
-        return GPTNeoConfig.from_pretrained("gpt_neo")
+        return GPTNeoConfig.from_pretrained("gpt-neo-125M")
 
     def prepare_config_and_inputs(self):
         input_ids = ids_tensor([self.batch_size, self.seq_length], self.vocab_size)

tests/test_modeling_flax_gptj.py → tests/gptj/test_modeling_flax_gptj.py

@@ -22,8 +22,8 @@ import transformers
 from transformers import GPT2Tokenizer, GPTJConfig, is_flax_available, is_torch_available
 from transformers.testing_utils import is_pt_flax_cross_test, require_flax, tooslow
 
-from .test_generation_flax_utils import FlaxGenerationTesterMixin
-from .test_modeling_flax_common import FlaxModelTesterMixin, ids_tensor, random_attention_mask
+from ..generation.test_generation_flax_utils import FlaxGenerationTesterMixin
+from ..test_modeling_flax_common import FlaxModelTesterMixin, ids_tensor, random_attention_mask
 
 
 if is_flax_available():

tests/test_modeling_gptj.py → tests/gptj/test_modeling_gptj.py

@@ -20,9 +20,9 @@ import unittest
 from transformers import GPTJConfig, is_torch_available
 from transformers.testing_utils import require_torch, slow, tooslow, torch_device
 
-from .test_configuration_common import ConfigTester
-from .test_generation_utils import GenerationTesterMixin
-from .test_modeling_common import ModelTesterMixin, floats_tensor, ids_tensor, random_attention_mask
+from ..generation.test_generation_utils import GenerationTesterMixin
+from ..test_configuration_common import ConfigTester
+from ..test_modeling_common import ModelTesterMixin, floats_tensor, ids_tensor, random_attention_mask
 
 
 if is_torch_available():

tests/test_tokenization_herbert.py → tests/herbert/test_tokenization_herbert.py

@@ -22,7 +22,7 @@ from transformers import HerbertTokenizer, HerbertTokenizerFast
 from transformers.models.herbert.tokenization_herbert import VOCAB_FILES_NAMES
 from transformers.testing_utils import get_tests_dir, require_tokenizers, slow
 
-from .test_tokenization_common import TokenizerTesterMixin
+from ..test_tokenization_common import TokenizerTesterMixin
 
 
 @require_tokenizers

tests/test_modeling_hubert.py → tests/hubert/test_modeling_hubert.py

@@ -20,12 +20,17 @@ import unittest
 
 import pytest
 
-from tests.test_modeling_common import floats_tensor, ids_tensor, random_attention_mask
 from transformers import HubertConfig, is_torch_available
 from transformers.testing_utils import require_soundfile, require_torch, slow, torch_device
 
-from .test_configuration_common import ConfigTester
-from .test_modeling_common import ModelTesterMixin, _config_zero_init
+from ..test_configuration_common import ConfigTester
+from ..test_modeling_common import (
+    ModelTesterMixin,
+    _config_zero_init,
+    floats_tensor,
+    ids_tensor,
+    random_attention_mask,
+)
 
 
 if is_torch_available():

tests/test_modeling_tf_hubert.py → tests/hubert/test_modeling_tf_hubert.py

@@ -25,8 +25,8 @@ import pytest
 from transformers import is_tf_available
 from transformers.testing_utils import require_soundfile, require_tf, slow
 
-from .test_configuration_common import ConfigTester
-from .test_modeling_tf_common import TFModelTesterMixin, ids_tensor
+from ..test_configuration_common import ConfigTester
+from ..test_modeling_tf_common import TFModelTesterMixin, ids_tensor
 
 
 if is_tf_available():

tests/test_modeling_ibert.py → tests/ibert/test_modeling_ibert.py

@@ -20,8 +20,8 @@ import unittest
 from transformers import IBertConfig, is_torch_available
 from transformers.testing_utils import require_torch, slow, torch_device
 
-from .test_configuration_common import ConfigTester
-from .test_modeling_common import ModelTesterMixin, ids_tensor, random_attention_mask
+from ..test_configuration_common import ConfigTester
+from ..test_modeling_common import ModelTesterMixin, ids_tensor, random_attention_mask
 
 
 if is_torch_available():

tests/test_feature_extraction_imagegpt.py → tests/imagegpt/test_feature_extraction_imagegpt.py

@@ -25,7 +25,7 @@ from datasets import load_dataset
 from transformers.file_utils import is_torch_available, is_vision_available
 from transformers.testing_utils import require_torch, require_vision, slow
 
-from .test_feature_extraction_common import FeatureExtractionSavingTestMixin
+from ..test_feature_extraction_common import FeatureExtractionSavingTestMixin
 
 
 if is_torch_available():

tests/test_modeling_imagegpt.py → tests/imagegpt/test_modeling_imagegpt.py

@@ -24,9 +24,15 @@ from transformers import ImageGPTConfig
 from transformers.file_utils import cached_property, is_torch_available, is_vision_available
 from transformers.testing_utils import require_torch, require_vision, slow, torch_device
 
-from .test_configuration_common import ConfigTester
-from .test_generation_utils import GenerationTesterMixin
-from .test_modeling_common import ModelTesterMixin, _config_zero_init, floats_tensor, ids_tensor, random_attention_mask
+from ..generation.test_generation_utils import GenerationTesterMixin
+from ..test_configuration_common import ConfigTester
+from ..test_modeling_common import (
+    ModelTesterMixin,
+    _config_zero_init,
+    floats_tensor,
+    ids_tensor,
+    random_attention_mask,
+)
 
 
 if is_torch_available():

tests/test_modeling_layoutlm.py → tests/layoutlm/test_modeling_layoutlm.py

@@ -19,8 +19,8 @@ import unittest
 from transformers import LayoutLMConfig, is_torch_available
 from transformers.testing_utils import require_torch, slow, torch_device
 
-from .test_configuration_common import ConfigTester
-from .test_modeling_common import ModelTesterMixin, ids_tensor
+from ..test_configuration_common import ConfigTester
+from ..test_modeling_common import ModelTesterMixin, ids_tensor
 
 
 if is_torch_available():

tests/test_modeling_tf_layoutlm.py → tests/layoutlm/test_modeling_tf_layoutlm.py

@@ -20,8 +20,8 @@ import numpy as np
 from transformers import LayoutLMConfig, is_tf_available
 from transformers.testing_utils import require_tf, slow
 
-from .test_configuration_common import ConfigTester
-from .test_modeling_tf_common import TFModelTesterMixin, ids_tensor
+from ..test_configuration_common import ConfigTester
+from ..test_modeling_tf_common import TFModelTesterMixin, ids_tensor
 
 
 if is_tf_available():

tests/test_tokenization_layoutlm.py → tests/layoutlm/test_tokenization_layoutlm.py

@@ -21,7 +21,7 @@ from transformers import LayoutLMTokenizer, LayoutLMTokenizerFast
 from transformers.models.layoutlm.tokenization_layoutlm import VOCAB_FILES_NAMES
 from transformers.testing_utils import require_tokenizers
 
-from .test_tokenization_common import TokenizerTesterMixin
+from ..test_tokenization_common import TokenizerTesterMixin
 
 
 @require_tokenizers