Speed up TF tests by reducing hidden layer counts (#24595)

* hidden layers, huh, what are they good for (absolutely nothing)

* Some tests break with 1 hidden layer, use 2

* Use 1 hidden layer in a few slow models

* Use num_hidden_layers=2 everywhere

* Slightly higher tol for groupvit

* Slightly higher tol for groupvit

tests/models/gptj/test_modeling_tf_gptj.py

@@ -51,7 +51,7 @@ class TFGPTJModelTester:
         self.vocab_size = 99
         self.hidden_size = 32
         self.rotary_dim = 4
-        self.num_hidden_layers = 5
+        self.num_hidden_layers = 2
         self.num_attention_heads = 4
         self.intermediate_size = 37
         self.hidden_act = "gelu"

tests/models/groupvit/test_modeling_tf_groupvit.py

@@ -150,6 +150,10 @@ class TFGroupViTVisionModelTest(TFModelTesterMixin, unittest.TestCase):
     test_head_masking = False
     test_onnx = False
 
+    def check_pt_tf_outputs(self, tf_outputs, pt_outputs, model_class, tol=1e-4, name="outputs", attributes=None):
+        # We override with a slightly higher tol value, as this model tends to diverge a bit more
+        super().check_pt_tf_outputs(tf_outputs, pt_outputs, model_class, tol, name, attributes)
+
     def setUp(self):
         self.model_tester = TFGroupViTVisionModelTester(self)
         self.config_tester = ConfigTester(
@@ -381,7 +385,7 @@ class TFGroupViTTextModelTester:
         use_labels=True,
         vocab_size=99,
         hidden_size=32,
-        num_hidden_layers=5,
+        num_hidden_layers=2,
         num_attention_heads=4,
         intermediate_size=37,
         dropout=0.1,
@@ -459,6 +463,10 @@ class TFGroupViTTextModelTest(TFModelTesterMixin, unittest.TestCase):
     test_head_masking = False
     test_onnx = False
 
+    def check_pt_tf_outputs(self, tf_outputs, pt_outputs, model_class, tol=1e-4, name="outputs", attributes=None):
+        # We override with a slightly higher tol value, as this model tends to diverge a bit more
+        super().check_pt_tf_outputs(tf_outputs, pt_outputs, model_class, tol, name, attributes)
+
     def setUp(self):
         self.model_tester = TFGroupViTTextModelTester(self)
         self.config_tester = ConfigTester(self, config_class=GroupViTTextConfig, hidden_size=37)
@@ -581,6 +589,10 @@ class TFGroupViTModelTest(TFModelTesterMixin, PipelineTesterMixin, unittest.Test
     test_attention_outputs = False
     test_onnx = False
 
+    def check_pt_tf_outputs(self, tf_outputs, pt_outputs, model_class, tol=1e-4, name="outputs", attributes=None):
+        # We override with a slightly higher tol value, as this model tends to diverge a bit more
+        super().check_pt_tf_outputs(tf_outputs, pt_outputs, model_class, tol, name, attributes)
+
     def setUp(self):
         self.model_tester = TFGroupViTModelTester(self)
 

tests/models/hubert/test_modeling_tf_hubert.py

@@ -59,7 +59,7 @@ class TFHubertModelTester:
         conv_bias=False,
         num_conv_pos_embeddings=16,
         num_conv_pos_embedding_groups=2,
-        num_hidden_layers=4,
+        num_hidden_layers=2,
         num_attention_heads=2,
         hidden_dropout_prob=0.1,  # this is most likely not correctly set yet
         intermediate_size=20,

tests/models/layoutlm/test_modeling_tf_layoutlm.py

@@ -52,7 +52,7 @@ class TFLayoutLMModelTester:
         use_labels=True,
         vocab_size=99,
         hidden_size=32,
-        num_hidden_layers=5,
+        num_hidden_layers=2,
         num_attention_heads=4,
         intermediate_size=37,
         hidden_act="gelu",

tests/models/layoutlmv3/test_modeling_tf_layoutlmv3.py

@@ -69,7 +69,7 @@ class TFLayoutLMv3ModelTester:
         use_labels=True,
         vocab_size=99,
         hidden_size=36,
-        num_hidden_layers=3,
+        num_hidden_layers=2,
         num_attention_heads=4,
         intermediate_size=37,
         hidden_act="gelu",

tests/models/led/test_modeling_tf_led.py

@@ -47,7 +47,7 @@ class TFLEDModelTester:
         use_labels=False,
         vocab_size=99,
         hidden_size=32,
-        num_hidden_layers=5,
+        num_hidden_layers=2,
         num_attention_heads=4,
         intermediate_size=37,
         hidden_dropout_prob=0.1,

tests/models/longformer/test_modeling_tf_longformer.py

@@ -56,7 +56,7 @@ class TFLongformerModelTester:
         self.use_labels = True
         self.vocab_size = 99
         self.hidden_size = 32
-        self.num_hidden_layers = 5
+        self.num_hidden_layers = 2
         self.num_attention_heads = 4
         self.intermediate_size = 37
         self.hidden_act = "gelu"

tests/models/marian/test_modeling_tf_marian.py

@@ -49,7 +49,7 @@ class TFMarianModelTester:
         use_labels=False,
         vocab_size=99,
         hidden_size=32,
-        num_hidden_layers=5,
+        num_hidden_layers=2,
         num_attention_heads=4,
         intermediate_size=37,
         hidden_dropout_prob=0.1,

tests/models/mbart/test_modeling_tf_mbart.py

@@ -47,7 +47,7 @@ class TFMBartModelTester:
         use_labels=False,
         vocab_size=99,
         hidden_size=32,
-        num_hidden_layers=5,
+        num_hidden_layers=2,
         num_attention_heads=4,
         intermediate_size=37,
         hidden_dropout_prob=0.1,

tests/models/mobilebert/test_modeling_tf_mobilebert.py

@@ -97,7 +97,7 @@ class TFMobileBertModelTest(TFModelTesterMixin, PipelineTesterMixin, unittest.Te
             vocab_size=99,
             hidden_size=32,
             embedding_size=32,
-            num_hidden_layers=5,
+            num_hidden_layers=2,
             num_attention_heads=4,
             intermediate_size=37,
             hidden_act="gelu",

tests/models/mpnet/test_modeling_tf_mpnet.py

@@ -51,7 +51,7 @@ class TFMPNetModelTester:
         use_labels=True,
         vocab_size=99,
         hidden_size=64,
-        num_hidden_layers=5,
+        num_hidden_layers=2,
         num_attention_heads=4,
         intermediate_size=64,
         hidden_act="gelu",

tests/models/openai/test_modeling_tf_openai.py

@@ -53,7 +53,7 @@ class TFOpenAIGPTModelTester:
         self.use_mc_token_ids = True
         self.vocab_size = 99
         self.hidden_size = 32
-        self.num_hidden_layers = 5
+        self.num_hidden_layers = 2
         self.num_attention_heads = 4
         self.intermediate_size = 37
         self.hidden_act = "gelu"

tests/models/pegasus/test_modeling_tf_pegasus.py

@@ -47,7 +47,7 @@ class TFPegasusModelTester:
         use_labels=False,
         vocab_size=99,
         hidden_size=32,
-        num_hidden_layers=5,
+        num_hidden_layers=2,
         num_attention_heads=4,
         intermediate_size=37,
         hidden_dropout_prob=0.1,

tests/models/rembert/test_modeling_tf_rembert.py

@@ -54,7 +54,7 @@ class TFRemBertModelTester:
         hidden_size=32,
         input_embedding_size=18,
         output_embedding_size=43,
-        num_hidden_layers=5,
+        num_hidden_layers=2,
         num_attention_heads=4,
         intermediate_size=37,
         hidden_act="gelu",
@@ -79,7 +79,7 @@ class TFRemBertModelTester:
         self.hidden_size = 32
         self.input_embedding_size = input_embedding_size
         self.output_embedding_size = output_embedding_size
-        self.num_hidden_layers = 5
+        self.num_hidden_layers = 2
         self.num_attention_heads = 4
         self.intermediate_size = 37
         self.hidden_act = "gelu"

tests/models/roberta/test_modeling_tf_roberta.py

@@ -56,7 +56,7 @@ class TFRobertaModelTester:
         self.use_labels = True
         self.vocab_size = 99
         self.hidden_size = 32
-        self.num_hidden_layers = 5
+        self.num_hidden_layers = 2
         self.num_attention_heads = 4
         self.intermediate_size = 37
         self.hidden_act = "gelu"

tests/models/roberta_prelayernorm/test_modeling_tf_roberta_prelayernorm.py

@@ -57,7 +57,7 @@ class TFRobertaPreLayerNormModelTester:
         self.use_labels = True
         self.vocab_size = 99
         self.hidden_size = 32
-        self.num_hidden_layers = 5
+        self.num_hidden_layers = 2
         self.num_attention_heads = 4
         self.intermediate_size = 37
         self.hidden_act = "gelu"

tests/models/roformer/test_modeling_tf_roformer.py

@@ -56,7 +56,7 @@ class TFRoFormerModelTester:
         use_labels=True,
         vocab_size=99,
         hidden_size=32,
-        num_hidden_layers=5,
+        num_hidden_layers=2,
         num_attention_heads=4,
         intermediate_size=37,
         hidden_act="gelu",
@@ -79,7 +79,7 @@ class TFRoFormerModelTester:
         self.use_labels = True
         self.vocab_size = 99
         self.hidden_size = 32
-        self.num_hidden_layers = 5
+        self.num_hidden_layers = 2
         self.num_attention_heads = 4
         self.intermediate_size = 37
         self.hidden_act = "gelu"

tests/models/t5/test_modeling_tf_t5.py

@@ -46,7 +46,7 @@ class TFT5ModelTester:
         self.vocab_size = 99
         self.n_positions = 14
         self.hidden_size = 32
-        self.num_hidden_layers = 5
+        self.num_hidden_layers = 2
         self.num_attention_heads = 4
         self.d_ff = 37
         self.relative_attention_num_buckets = 8
@@ -325,7 +325,7 @@ class TFT5EncoderOnlyModelTester:
         # For common tests
         use_attention_mask=True,
         hidden_size=32,
-        num_hidden_layers=5,
+        num_hidden_layers=2,
         num_attention_heads=4,
         d_ff=37,
         relative_attention_num_buckets=8,

tests/models/tapas/test_modeling_tf_tapas.py

@@ -77,7 +77,7 @@ class TFTapasModelTester:
         use_labels=True,
         vocab_size=99,
         hidden_size=32,
-        num_hidden_layers=5,
+        num_hidden_layers=2,
         num_attention_heads=4,
         intermediate_size=37,
         hidden_act="gelu",

tests/models/transfo_xl/test_modeling_tf_transfo_xl.py

@@ -59,7 +59,7 @@ class TFTransfoXLModelTester:
         self.d_head = 8
         self.d_inner = 128
         self.div_val = 2
-        self.num_hidden_layers = 5
+        self.num_hidden_layers = 2
         self.scope = None
         self.seed = 1
         self.eos_token_id = 0