from_pretrained: check that the pretrained model is for the right model architecture (#10586)

* Added check to ensure model name passed to from_pretrained and model are the same

* Added test to check from_pretrained throws assert error when passed an incompatiable model name

* Modified assert in from_pretrained with f-strings. Modified test to ensure desired assert message is being generated

* Added check to ensure config and model has model_type

* Fix FlauBERT heads

Co-authored-by: vimarsh chaturvedi <vimarsh chaturvedi>
Co-authored-by: Stas Bekman <stas@stason.org>
Co-authored-by: Lysandre <lysandre.debut@reseau.eseo.fr>

src/transformers/configuration_utils.py

@@ -384,6 +384,11 @@ class PretrainedConfig(object):
 
         """
         config_dict, kwargs = cls.get_config_dict(pretrained_model_name_or_path, **kwargs)
+        if config_dict.get("model_type", False) and hasattr(cls, "model_type"):
+            assert (
+                config_dict["model_type"] == cls.model_type
+            ), f"You tried to initiate a model of type '{cls.model_type}' with a pretrained model of type '{config_dict['model_type']}'"
+
         return cls.from_dict(config_dict, **kwargs)
 
     @classmethod

src/transformers/models/flaubert/modeling_tf_flaubert.py

@@ -932,6 +932,8 @@ class TFFlaubertForQuestionAnsweringSimple(TFXLMForQuestionAnsweringSimple):
     FLAUBERT_START_DOCSTRING,
 )
 class TFFlaubertForTokenClassification(TFXLMForTokenClassification):
+    config_class = FlaubertConfig
+
     def __init__(self, config, *inputs, **kwargs):
         super().__init__(config, *inputs, **kwargs)
         self.transformer = TFFlaubertMainLayer(config, name="transformer")
@@ -945,6 +947,8 @@ class TFFlaubertForTokenClassification(TFXLMForTokenClassification):
     FLAUBERT_START_DOCSTRING,
 )
 class TFFlaubertForMultipleChoice(TFXLMForMultipleChoice):
+    config_class = FlaubertConfig
+
     def __init__(self, config, *inputs, **kwargs):
         super().__init__(config, *inputs, **kwargs)
         self.transformer = TFFlaubertMainLayer(config, name="transformer")

tests/test_modeling_common.py

@@ -47,6 +47,7 @@ if is_torch_available():
         BertModel,
         PretrainedConfig,
         PreTrainedModel,
+        T5ForConditionalGeneration,
     )
 
 
@@ -58,6 +59,9 @@ def _config_zero_init(config):
     return configs_no_init
 
 
+TINY_T5 = "patrickvonplaten/t5-tiny-random"
+
+
 @require_torch
 class ModelTesterMixin:
 
@@ -1284,3 +1288,11 @@ class ModelUtilsTest(unittest.TestCase):
             model = BertModel.from_pretrained(model_name, output_attentions=True, output_hidden_states=True)
             self.assertEqual(model.config.output_hidden_states, True)
             self.assertEqual(model.config, config)
+
+    def test_model_from_pretrained_with_different_pretrained_model_name(self):
+        model = T5ForConditionalGeneration.from_pretrained(TINY_T5)
+        self.assertIsNotNone(model)
+
+        with self.assertRaises(Exception) as context:
+            BertModel.from_pretrained(TINY_T5)
+        self.assertTrue("You tried to initiate a model of type" in str(context.exception))