Update tiny model summary file (#27388)

* update

* fix

---------

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

tests/models/clvp/test_modeling_clvp.py

@@ -38,6 +38,7 @@ from ...test_modeling_common import (
     ids_tensor,
     random_attention_mask,
 )
+from ...test_pipeline_mixin import PipelineTesterMixin
 
 
 if is_torch_available():
@@ -281,9 +282,10 @@ class ClvpDecoderTester:
 
 
 @require_torch
-class ClvpDecoderTest(ModelTesterMixin, GenerationTesterMixin, unittest.TestCase):
+class ClvpDecoderTest(ModelTesterMixin, GenerationTesterMixin, PipelineTesterMixin, unittest.TestCase):
     all_model_classes = (ClvpModel, ClvpForCausalLM) if is_torch_available() else ()
     all_generative_model_classes = (ClvpForCausalLM,) if is_torch_available() else ()
+    pipeline_model_mapping = {"feature-extraction": ClvpModelForConditionalGeneration} if is_torch_available() else {}
 
     test_pruning = False
 

tests/models/fuyu/test_modeling_fuyu.py

@@ -24,6 +24,7 @@ from transformers.testing_utils import require_torch, require_torch_gpu, slow, t
 from transformers.utils import cached_property
 
 from ...test_modeling_common import ModelTesterMixin, ids_tensor, random_attention_mask
+from ...test_pipeline_mixin import PipelineTesterMixin
 
 
 if is_vision_available():
@@ -262,9 +263,9 @@ class FuyuModelTester:
 
 
 @require_torch
-class FuyuModelTest(ModelTesterMixin, unittest.TestCase):
+class FuyuModelTest(ModelTesterMixin, PipelineTesterMixin, unittest.TestCase):
     all_model_classes = (FuyuForCausalLM,) if is_torch_available() else ()
-    pipeline_model_mapping = {"image-to-text": FuyuForCausalLM} if is_torch_available() else {}
+    pipeline_model_mapping = {"text-generation": FuyuForCausalLM} if is_torch_available() else {}
 
     test_head_masking = False
     test_pruning = False

tests/models/kosmos2/test_modeling_kosmos2.py

@@ -37,6 +37,7 @@ from ...test_modeling_common import (
     ids_tensor,
     random_attention_mask,
 )
+from ...test_pipeline_mixin import PipelineTesterMixin
 
 
 if is_torch_available():
@@ -244,15 +245,26 @@ class Kosmos2ModelTester:
 
 
 @require_torch
-class Kosmos2ModelTest(ModelTesterMixin, unittest.TestCase):
+class Kosmos2ModelTest(ModelTesterMixin, PipelineTesterMixin, unittest.TestCase):
     all_model_classes = (Kosmos2Model, Kosmos2ForConditionalGeneration) if is_torch_available() else ()
     all_generative_model_classes = (Kosmos2ForConditionalGeneration,) if is_torch_available() else ()
+    pipeline_model_mapping = (
+        {"feature-extraction": Kosmos2Model, "image-to-text": Kosmos2ForConditionalGeneration}
+        if is_torch_available()
+        else {}
+    )
     fx_compatible = False
     test_head_masking = False
     test_pruning = False
     test_resize_embeddings = False
     test_attention_outputs = False
 
+    # TODO: `image-to-text` pipeline for this model needs Processor.
+    def is_pipeline_test_to_skip(
+        self, pipeline_test_casse_name, config_class, model_architecture, tokenizer_name, processor_name
+    ):
+        return pipeline_test_casse_name == "ImageToTextPipelineTests"
+
     def _prepare_for_class(self, inputs_dict, model_class, return_labels=False):
         inputs_dict = copy.deepcopy(inputs_dict)
 

tests/models/seamless_m4t/test_modeling_seamless_m4t.py

@@ -34,6 +34,7 @@ from ...test_modeling_common import (
     ids_tensor,
     random_attention_mask,
 )
+from ...test_pipeline_mixin import PipelineTesterMixin
 
 
 if is_torch_available():
@@ -616,7 +617,9 @@ class SeamlessM4TModelWithSpeechInputTest(ModelTesterMixin, unittest.TestCase):
 
 
 @require_torch
-class SeamlessM4TModelWithTextInputTest(ModelTesterMixin, GenerationTesterMixin, unittest.TestCase):
+class SeamlessM4TModelWithTextInputTest(
+    ModelTesterMixin, GenerationTesterMixin, PipelineTesterMixin, unittest.TestCase
+):
     is_encoder_decoder = True
     fx_compatible = False
     test_missing_keys = False
@@ -636,6 +639,19 @@ class SeamlessM4TModelWithTextInputTest(ModelTesterMixin, GenerationTesterMixin,
         else ()
     )
     all_generative_model_classes = (SeamlessM4TForTextToText,) if is_torch_available() else ()
+    pipeline_model_mapping = (
+        {
+            "automatic-speech-recognition": SeamlessM4TForSpeechToText,
+            "conversational": SeamlessM4TForTextToText,
+            "feature-extraction": SeamlessM4TModel,
+            "summarization": SeamlessM4TForTextToText,
+            "text-to-audio": SeamlessM4TForTextToSpeech,
+            "text2text-generation": SeamlessM4TForTextToText,
+            "translation": SeamlessM4TForTextToText,
+        }
+        if is_torch_available()
+        else {}
+    )
 
     def setUp(self):
         self.model_tester = SeamlessM4TModelTester(self, input_modality="text")

tests/models/swin2sr/test_modeling_swin2sr.py

@@ -162,7 +162,11 @@ class Swin2SRModelTester:
 @require_torch
 class Swin2SRModelTest(ModelTesterMixin, PipelineTesterMixin, unittest.TestCase):
     all_model_classes = (Swin2SRModel, Swin2SRForImageSuperResolution) if is_torch_available() else ()
-    pipeline_model_mapping = {"feature-extraction": Swin2SRModel} if is_torch_available() else {}
+    pipeline_model_mapping = (
+        {"feature-extraction": Swin2SRModel, "image-to-image": Swin2SRForImageSuperResolution}
+        if is_torch_available()
+        else {}
+    )
 
     fx_compatible = False
     test_pruning = False

tests/models/whisper/test_modeling_whisper.py

@@ -367,6 +367,7 @@ class WhisperModelTest(ModelTesterMixin, GenerationTesterMixin, PipelineTesterMi
             "audio-classification": WhisperForAudioClassification,
             "automatic-speech-recognition": WhisperForConditionalGeneration,
             "feature-extraction": WhisperModel,
+            "text-generation": WhisperForCausalLM,
         }
         if is_torch_available()
         else {}