Check model inputs - hidden states (huggingface#40994)

* update all models

* fix copies

* skip aria tests

* update other models

* skip should be in test, not tester

* i think this is more descriptive as a name

* find and replace for new models

Author: zucchini-nlp

examples/modular-transformers/modeling_dummy_bert.py

@@ -657,7 +657,7 @@ class PreTrainedModel
         for layer, heads in heads_to_prune.items():
             self.encoder.layer[layer].attention.prune_heads(heads)
 
-    @check_model_inputs
+    @check_model_inputs()
     @auto_docstring
     def forward(
         self,

examples/modular-transformers/modeling_roberta.py

@@ -660,7 +660,7 @@ class PreTrainedModel
         for layer, heads in heads_to_prune.items():
             self.encoder.layer[layer].attention.prune_heads(heads)
 
-    @check_model_inputs
+    @check_model_inputs()
     @auto_docstring
     def forward(
         self,

examples/modular-transformers/modeling_super.py

@@ -323,7 +323,7 @@ def __init__(self, config: SuperConfig):
         # Initialize weights and apply final processing
         self.post_init()
 
-    @check_model_inputs
+    @check_model_inputs()
     @auto_docstring
     def forward(
         self,

src/transformers/models/aimv2/modeling_aimv2.py

@@ -444,7 +444,7 @@ def get_input_embeddings(self) -> nn.Module:
         return self.embeddings.patch_embed
 
     @deprecate_kwarg("attention_mask", version="v4.58.0")
-    @check_model_inputs
+    @check_model_inputs(tie_last_hidden_states=False)
     @auto_docstring
     def forward(
         self,
@@ -520,7 +520,7 @@ def get_input_embeddings(self) -> nn.Module:
     def set_input_embeddings(self, value):
         self.embeddings.token_embedding = value
 
-    @check_model_inputs
+    @check_model_inputs(tie_last_hidden_states=False)
     @auto_docstring
     def forward(
         self,

src/transformers/models/aimv2/modular_aimv2.py

@@ -488,7 +488,7 @@ def get_input_embeddings(self) -> nn.Module:
         return self.embeddings.patch_embed
 
     @deprecate_kwarg("attention_mask", version="v4.58.0")
-    @check_model_inputs
+    @check_model_inputs(tie_last_hidden_states=False)
     @auto_docstring
     def forward(
         self,
@@ -564,7 +564,7 @@ def get_input_embeddings(self) -> nn.Module:
     def set_input_embeddings(self, value):
         self.embeddings.token_embedding = value
 
-    @check_model_inputs
+    @check_model_inputs(tie_last_hidden_states=False)
     @auto_docstring
     def forward(
         self,

src/transformers/models/albert/modeling_albert.py

@@ -457,7 +457,7 @@ def _prune_heads(self, heads_to_prune: dict[int, list[int]]) -> None:
             inner_group_idx = int(layer - group_idx * self.config.inner_group_num)
             self.encoder.albert_layer_groups[group_idx].albert_layers[inner_group_idx].attention.prune_heads(heads)
 
-    @check_model_inputs
+    @check_model_inputs()
     @auto_docstring
     def forward(
         self,

src/transformers/models/apertus/modeling_apertus.py

@@ -339,7 +339,7 @@ def __init__(self, config: ApertusConfig):
         # Initialize weights and apply final processing
         self.post_init()
 
-    @check_model_inputs
+    @check_model_inputs()
     @auto_docstring
     def forward(
         self,

src/transformers/models/arcee/modeling_arcee.py

@@ -344,7 +344,7 @@ def __init__(self, config: ArceeConfig):
         # Initialize weights and apply final processing
         self.post_init()
 
-    @check_model_inputs
+    @check_model_inputs()
     @auto_docstring
     def forward(
         self,

src/transformers/models/aria/modeling_aria.py

@@ -669,7 +669,7 @@ def __init__(self, config: AriaTextConfig):
         # Initialize weights and apply final processing
         self.post_init()
 
-    @check_model_inputs
+    @check_model_inputs()
     @auto_docstring
     def forward(
         self,

src/transformers/models/audio_spectrogram_transformer/modeling_audio_spectrogram_transformer.py

@@ -363,7 +363,7 @@ class PreTrainedModel
         for layer, heads in heads_to_prune.items():
             self.encoder.layer[layer].attention.prune_heads(heads)
 
-    @check_model_inputs
+    @check_model_inputs()
     @auto_docstring
     def forward(
         self,