Update quality tooling for formatting (#21480)

* Result of black 23.1

* Update target to Python 3.7

* Switch flake8 to ruff

* Configure isort

* Configure isort

* Apply isort with line limit

* Put the right black version

* adapt black in check copies

* Fix copies

tests/models/encoder_decoder/test_modeling_encoder_decoder.py

@@ -72,7 +72,7 @@ class EncoderDecoderMixin:
         decoder_config,
         decoder_input_ids,
         decoder_attention_mask,
-        **kwargs
+        **kwargs,
     ):
         encoder_decoder_config = EncoderDecoderConfig.from_encoder_decoder_configs(config, decoder_config)
         self.assertTrue(encoder_decoder_config.decoder.is_decoder)
@@ -106,7 +106,7 @@ class EncoderDecoderMixin:
         decoder_config,
         decoder_input_ids,
         decoder_attention_mask,
-        **kwargs
+        **kwargs,
     ):
         encoder_model, decoder_model = self.get_encoder_decoder_model(config, decoder_config)
         enc_dec_model = EncoderDecoderModel(encoder=encoder_model, decoder=decoder_model)
@@ -167,7 +167,7 @@ class EncoderDecoderMixin:
         decoder_config,
         decoder_input_ids,
         decoder_attention_mask,
-        **kwargs
+        **kwargs,
     ):
         encoder_model, decoder_model = self.get_encoder_decoder_model(config, decoder_config)
         with tempfile.TemporaryDirectory() as encoder_tmp_dirname, tempfile.TemporaryDirectory() as decoder_tmp_dirname:
@@ -210,7 +210,7 @@ class EncoderDecoderMixin:
         decoder_input_ids,
         decoder_attention_mask,
         return_dict,
-        **kwargs
+        **kwargs,
     ):
         encoder_model, decoder_model = self.get_encoder_decoder_model(config, decoder_config)
         kwargs = {"encoder_model": encoder_model, "decoder_model": decoder_model, "return_dict": return_dict}
@@ -240,7 +240,7 @@ class EncoderDecoderMixin:
         decoder_config,
         decoder_input_ids,
         decoder_attention_mask,
-        **kwargs
+        **kwargs,
     ):
         encoder_model, decoder_model = self.get_encoder_decoder_model(config, decoder_config)
         enc_dec_model = EncoderDecoderModel(encoder=encoder_model, decoder=decoder_model)
@@ -281,7 +281,7 @@ class EncoderDecoderMixin:
         decoder_config,
         decoder_input_ids,
         decoder_attention_mask,
-        **kwargs
+        **kwargs,
     ):
         encoder_model, decoder_model = self.get_encoder_decoder_model(config, decoder_config)
         enc_dec_model = EncoderDecoderModel(encoder=encoder_model, decoder=decoder_model)
@@ -327,7 +327,7 @@ class EncoderDecoderMixin:
         decoder_input_ids,
         decoder_attention_mask,
         labels,
-        **kwargs
+        **kwargs,
     ):
         encoder_model, decoder_model = self.get_encoder_decoder_model(config, decoder_config)
         enc_dec_model = EncoderDecoderModel(encoder=encoder_model, decoder=decoder_model)
@@ -395,7 +395,7 @@ class EncoderDecoderMixin:
         decoder_input_ids,
         decoder_attention_mask,
         labels,
-        **kwargs
+        **kwargs,
     ):
         # make the decoder inputs a different shape from the encoder inputs to harden the test
         decoder_input_ids = decoder_input_ids[:, :-1]
@@ -424,7 +424,7 @@ class EncoderDecoderMixin:
         decoder_input_ids,
         decoder_attention_mask,
         labels,
-        **kwargs
+        **kwargs,
     ):
         # Similar to `check_encoder_decoder_model_output_attentions`, but with `output_attentions` triggered from the
         # config file. Contrarily to most models, changing the model's config won't work -- the defaults are loaded
@@ -491,7 +491,7 @@ class EncoderDecoderMixin:
         decoder_input_ids,
         decoder_attention_mask,
         labels,
-        **kwargs
+        **kwargs,
     ):
         torch.manual_seed(0)
         encoder_model, decoder_model = self.get_encoder_decoder_model(config, decoder_config)

tests/models/encoder_decoder/test_modeling_flax_encoder_decoder.py

@@ -69,7 +69,7 @@ class FlaxEncoderDecoderMixin:
         decoder_config,
         decoder_input_ids,
         decoder_attention_mask,
-        **kwargs
+        **kwargs,
     ):
         encoder_decoder_config = EncoderDecoderConfig.from_encoder_decoder_configs(config, decoder_config)
         self.assertTrue(encoder_decoder_config.decoder.is_decoder)
@@ -102,7 +102,7 @@ class FlaxEncoderDecoderMixin:
         decoder_input_ids,
         decoder_attention_mask,
         return_dict,
-        **kwargs
+        **kwargs,
     ):
         encoder_model, decoder_model = self.get_encoder_decoder_model(config, decoder_config)
         kwargs = {"encoder_model": encoder_model, "decoder_model": decoder_model, "return_dict": return_dict}
@@ -131,7 +131,7 @@ class FlaxEncoderDecoderMixin:
         decoder_config,
         decoder_input_ids,
         decoder_attention_mask,
-        **kwargs
+        **kwargs,
     ):
         encoder_model, decoder_model = self.get_encoder_decoder_model(config, decoder_config)
         kwargs = {"encoder_model": encoder_model, "decoder_model": decoder_model}
@@ -170,7 +170,7 @@ class FlaxEncoderDecoderMixin:
         decoder_config,
         decoder_input_ids,
         decoder_attention_mask,
-        **kwargs
+        **kwargs,
     ):
         encoder_model, decoder_model = self.get_encoder_decoder_model(config, decoder_config)
         # assert that model attributes match those of configs
@@ -215,7 +215,7 @@ class FlaxEncoderDecoderMixin:
         decoder_config,
         decoder_input_ids,
         decoder_attention_mask,
-        **kwargs
+        **kwargs,
     ):
         # make the decoder inputs a different shape from the encoder inputs to harden the test
         decoder_input_ids = decoder_input_ids[:, :-1]
@@ -292,7 +292,6 @@ class FlaxEncoderDecoderMixin:
         self.assertEqual(generated_sequences.shape, (input_ids.shape[0],) + (decoder_config.max_length,))
 
     def check_pt_flax_equivalence(self, pt_model, fx_model, inputs_dict):
-
         pt_model.to(torch_device)
         pt_model.eval()
 
@@ -334,7 +333,6 @@ class FlaxEncoderDecoderMixin:
             self.assert_almost_equals(fx_output, pt_output_loaded.numpy(), 1e-5)
 
     def check_equivalence_pt_to_flax(self, config, decoder_config, inputs_dict):
-
         encoder_decoder_config = EncoderDecoderConfig.from_encoder_decoder_configs(config, decoder_config)
 
         pt_model = EncoderDecoderModel(encoder_decoder_config)
@@ -346,7 +344,6 @@ class FlaxEncoderDecoderMixin:
         self.check_pt_flax_equivalence(pt_model, fx_model, inputs_dict)
 
     def check_equivalence_flax_to_pt(self, config, decoder_config, inputs_dict):
-
         encoder_decoder_config = EncoderDecoderConfig.from_encoder_decoder_configs(config, decoder_config)
 
         pt_model = EncoderDecoderModel(encoder_decoder_config)
@@ -390,7 +387,6 @@ class FlaxEncoderDecoderMixin:
 
     @is_pt_flax_cross_test
     def test_pt_flax_equivalence(self):
-
         config_inputs_dict = self.prepare_config_and_inputs()
         config = config_inputs_dict.pop("config")
         decoder_config = config_inputs_dict.pop("decoder_config")
@@ -589,7 +585,6 @@ class FlaxEncoderDecoderModelTest(unittest.TestCase):
         return FlaxEncoderDecoderModel.from_encoder_decoder_pretrained("bert-base-cased", "gpt2")
 
     def _check_configuration_tie(self, model):
-
         module = model.module.bind(model.params)
 
         assert id(module.decoder.config) == id(model.config.decoder)

tests/models/encoder_decoder/test_modeling_tf_encoder_decoder.py

@@ -78,7 +78,7 @@ class TFEncoderDecoderMixin:
         decoder_config,
         decoder_input_ids,
         decoder_attention_mask,
-        **kwargs
+        **kwargs,
     ):
         encoder_decoder_config = EncoderDecoderConfig.from_encoder_decoder_configs(config, decoder_config)
         self.assertTrue(encoder_decoder_config.decoder.is_decoder)
@@ -111,7 +111,7 @@ class TFEncoderDecoderMixin:
         decoder_config,
         decoder_input_ids,
         decoder_attention_mask,
-        **kwargs
+        **kwargs,
     ):
         encoder_model, decoder_model = self.get_encoder_decoder_model(config, decoder_config)
         enc_dec_model = TFEncoderDecoderModel(encoder=encoder_model, decoder=decoder_model)
@@ -160,7 +160,7 @@ class TFEncoderDecoderMixin:
         decoder_input_ids,
         decoder_attention_mask,
         return_dict,
-        **kwargs
+        **kwargs,
     ):
         encoder_model, decoder_model = self.get_encoder_decoder_model(config, decoder_config)
         kwargs = {"encoder_model": encoder_model, "decoder_model": decoder_model, "return_dict": return_dict}
@@ -190,7 +190,7 @@ class TFEncoderDecoderMixin:
         decoder_config,
         decoder_input_ids,
         decoder_attention_mask,
-        **kwargs
+        **kwargs,
     ):
         encoder_model, decoder_model = self.get_encoder_decoder_model(config, decoder_config)
         enc_dec_model = TFEncoderDecoderModel(encoder=encoder_model, decoder=decoder_model)
@@ -231,7 +231,7 @@ class TFEncoderDecoderMixin:
         decoder_input_ids,
         decoder_attention_mask,
         labels,
-        **kwargs
+        **kwargs,
     ):
         encoder_model, decoder_model = self.get_encoder_decoder_model(config, decoder_config)
         enc_dec_model = TFEncoderDecoderModel(encoder=encoder_model, decoder=decoder_model)
@@ -298,7 +298,7 @@ class TFEncoderDecoderMixin:
         decoder_config,
         decoder_input_ids,
         decoder_attention_mask,
-        **kwargs
+        **kwargs,
     ):
         # make the decoder inputs a different shape from the encoder inputs to harden the test
         decoder_input_ids = decoder_input_ids[:, :-1]
@@ -326,7 +326,7 @@ class TFEncoderDecoderMixin:
         decoder_config,
         decoder_input_ids,
         decoder_attention_mask,
-        **kwargs
+        **kwargs,
     ):
         # Similar to `check_encoder_decoder_model_output_attentions`, but with `output_attentions` triggered from the
         # config file. Contrarily to most models, changing the model's config won't work -- the defaults are loaded
@@ -470,7 +470,6 @@ class TFEncoderDecoderMixin:
             )
 
     def prepare_pt_inputs_from_tf_inputs(self, tf_inputs_dict):
-
         pt_inputs_dict = {}
         for name, key in tf_inputs_dict.items():
             if type(key) == bool:
@@ -490,7 +489,6 @@ class TFEncoderDecoderMixin:
         return pt_inputs_dict
 
     def check_pt_tf_models(self, tf_model, pt_model, tf_inputs_dict):
-
         pt_inputs_dict = self.prepare_pt_inputs_from_tf_inputs(tf_inputs_dict)
 
         # send pytorch inputs to the correct device
@@ -607,7 +605,6 @@ class TFEncoderDecoderMixin:
 
     @is_pt_tf_cross_test
     def test_pt_tf_model_equivalence(self):
-
         config_inputs_dict = self.prepare_config_and_inputs()
         labels = config_inputs_dict.pop("decoder_token_labels")
 
@@ -762,7 +759,6 @@ class TFBertEncoderDecoderModelTest(TFEncoderDecoderMixin, unittest.TestCase):
     @slow
     @is_pt_tf_cross_test
     def test_bert2bert_summarization(self):
-
         from transformers import EncoderDecoderModel
 
         tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
@@ -863,7 +859,6 @@ class TFGPT2EncoderDecoderModelTest(TFEncoderDecoderMixin, unittest.TestCase):
     @slow
     @is_pt_tf_cross_test
     def test_bert2gpt2_summarization(self):
-
         from transformers import EncoderDecoderModel
 
         tokenizer_in = AutoTokenizer.from_pretrained("bert-base-cased")
@@ -1171,7 +1166,6 @@ class TFEncoderDecoderModelSaveLoadTests(unittest.TestCase):
         decoder_input_ids = decoder_tokenizer("Linda Davis", return_tensors="tf").input_ids
 
         with tempfile.TemporaryDirectory() as tmp_dirname:
-
             # Since most of HF's models don't have pretrained cross-attention layers, they are randomly
             # initialized even if we create models using `from_pretrained` method.
             # For the tests, the decoder need to be a model with pretrained cross-attention layers.