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/albert/test_modeling_tf_albert.py

@@ -56,7 +56,7 @@ class TFAlbertModelTester:
         vocab_size=99,
         embedding_size=16,
         hidden_size=32,
-        num_hidden_layers=5,
+        num_hidden_layers=2,
         num_attention_heads=4,
         intermediate_size=37,
         hidden_act="gelu",
@@ -80,7 +80,7 @@ class TFAlbertModelTester:
         self.vocab_size = 99
         self.embedding_size = 16
         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/bart/test_modeling_tf_bart.py

@@ -52,7 +52,7 @@ class TFBartModelTester:
         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/bert/test_modeling_tf_bert.py

@@ -57,7 +57,7 @@ class TFBertModelTester:
         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",
@@ -80,7 +80,7 @@ class TFBertModelTester:
         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/blenderbot/test_modeling_tf_blenderbot.py

@@ -48,7 +48,7 @@ class TFBlenderbotModelTester:
         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/blenderbot_small/test_modeling_tf_blenderbot_small.py

@@ -48,7 +48,7 @@ class TFBlenderbotSmallModelTester:
         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/blip/test_modeling_tf_blip.py

@@ -64,7 +64,7 @@ class TFBlipVisionModelTester:
         is_training=True,
         hidden_size=32,
         projection_dim=32,
-        num_hidden_layers=5,
+        num_hidden_layers=2,
         num_attention_heads=4,
         intermediate_size=37,
         dropout=0.1,
@@ -207,7 +207,7 @@ class TFBlipTextModelTester:
         vocab_size=99,
         hidden_size=32,
         projection_dim=32,
-        num_hidden_layers=5,
+        num_hidden_layers=2,
         num_attention_heads=4,
         intermediate_size=37,
         dropout=0.1,

tests/models/blip/test_modeling_tf_blip_text.py

@@ -46,7 +46,7 @@ class BlipTextModelTester:
         vocab_size=99,
         hidden_size=32,
         projection_dim=32,
-        num_hidden_layers=5,
+        num_hidden_layers=2,
         num_attention_heads=4,
         intermediate_size=37,
         dropout=0.1,

tests/models/clip/test_modeling_tf_clip.py

@@ -57,7 +57,7 @@ class TFCLIPVisionModelTester:
         num_channels=3,
         is_training=True,
         hidden_size=32,
-        num_hidden_layers=5,
+        num_hidden_layers=2,
         num_attention_heads=4,
         intermediate_size=37,
         dropout=0.1,
@@ -328,7 +328,7 @@ class TFCLIPTextModelTester:
         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,

tests/models/convbert/test_modeling_tf_convbert.py

@@ -51,7 +51,7 @@ class TFConvBertModelTester:
         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",
@@ -74,7 +74,7 @@ class TFConvBertModelTester:
         self.use_labels = True
         self.vocab_size = 99
         self.hidden_size = 384
-        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/ctrl/test_modeling_tf_ctrl.py

@@ -52,7 +52,7 @@ class TFCTRLModelTester(object):
         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/data2vec/test_modeling_tf_data2vec_vision.py

@@ -61,7 +61,7 @@ class TFData2VecVisionModelTester:
         is_training=True,
         use_labels=True,
         hidden_size=32,
-        num_hidden_layers=4,
+        num_hidden_layers=2,
         num_attention_heads=4,
         intermediate_size=37,
         hidden_act="gelu",

tests/models/deberta/test_modeling_tf_deberta.py

@@ -50,7 +50,7 @@ class TFDebertaModelTester:
         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",
@@ -73,7 +73,7 @@ class TFDebertaModelTester:
         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/deberta_v2/test_modeling_tf_deberta_v2.py

@@ -50,7 +50,7 @@ class TFDebertaV2ModelTester:
         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/deit/test_modeling_tf_deit.py

@@ -60,7 +60,7 @@ class TFDeiTModelTester:
         is_training=True,
         use_labels=True,
         hidden_size=32,
-        num_hidden_layers=5,
+        num_hidden_layers=2,
         num_attention_heads=4,
         intermediate_size=37,
         hidden_act="gelu",

tests/models/distilbert/test_modeling_tf_distilbert.py

@@ -54,7 +54,7 @@ class TFDistilBertModelTester:
         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/dpr/test_modeling_tf_dpr.py

@@ -53,7 +53,7 @@ class TFDPRModelTester:
         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/electra/test_modeling_tf_electra.py

@@ -54,7 +54,7 @@ class TFElectraModelTester:
         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/esm/test_modeling_tf_esm.py

@@ -53,7 +53,7 @@ class TFEsmModelTester:
         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/flaubert/test_modeling_tf_flaubert.py

@@ -61,7 +61,7 @@ class TFFlaubertModelTester:
         self.vocab_size = 99
         self.n_special = 0
         self.hidden_size = 32
-        self.num_hidden_layers = 5
+        self.num_hidden_layers = 2
         self.num_attention_heads = 4
         self.hidden_dropout_prob = 0.1
         self.attention_probs_dropout_prob = 0.1

tests/models/gpt2/test_modeling_tf_gpt2.py

@@ -55,7 +55,7 @@ class TFGPT2ModelTester:
         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"