update multimodal doc and model names (Transchex)

docs/source/networks.rst

@@ -500,9 +500,9 @@ Nets
 .. autoclass:: Critic
   :members:
 
-`VLTransformers`
+`Transchex`
 ~~~~~~~~~~~~~~~~
-.. autoclass:: VLTransformers
+.. autoclass:: Transchex
   :members:
 
 `NetAdapter`

monai/networks/nets/__init__.py

@@ -80,17 +80,9 @@
     seresnext101,
 )
 from .torchvision_fc import TorchVisionFCModel, TorchVisionFullyConvModel
+from .transchex import BertAttention, BertMixedLayer, BertOutput, BertPreTrainedModel, MultiModal, Pooler, Transchex
 from .unet import UNet, Unet, unet
 from .unetr import UNETR
 from .varautoencoder import VarAutoEncoder
 from .vit import ViT
-from .vltransformer import (
-    BertAttention,
-    BertMixedLayer,
-    BertOutput,
-    BertPreTrainedModel,
-    MultiModal,
-    Pooler,
-    VLTransformers,
-)
 from .vnet import VNet

monai/networks/nets/vltransformer.py → monai/networks/nets/transchex.py

@@ -34,7 +34,7 @@
     "BertMixedLayer",
     "Pooler",
     "MultiModal",
-    "VLTransformers",
+    "Transchex",
 ]
 
 
@@ -266,9 +266,10 @@ def forward(self, input_ids, token_type_ids=None, vision_feats=None, attention_m
         return hidden_state_mixed
 
 
-class VLTransformers(torch.nn.Module):
+class Transchex(torch.nn.Module):
     """
-    Vision Language Multimodal Transformers
+    TransChex based on: "Hatamizadeh et al.,TransCheX: Self-Supervised Pretraining of Vision-Language
+    Transformers for Chest X-ray Analysis"
     """
 
     def __init__(
@@ -321,7 +322,7 @@ def __init__(
 
             # for 3-channel with image size of (224,224), patch size of (32,32), 3 classes, 2 language layers,
             # 2 vision layers, 2 mixed modality layers and dropout of 0.2 in the classification head
-            net = VLTransformers(in_channels=3,
+            net = Transchex(in_channels=3,
                                  img_size=(224, 224),
                                  num_classes=3,
                                  num_language_layers=2,
@@ -330,7 +331,7 @@ def __init__(
                                  drop_out=0.2)
 
         """
-        super(VLTransformers, self).__init__()
+        super(Transchex, self).__init__()
         bert_config = {
             "attention_probs_dropout_prob": attention_probs_dropout_prob,
             "classifier_dropout": None,

tests/min_tests.py

@@ -140,7 +140,7 @@ def run_testsuit():
         "test_zoom",
         "test_zoom_affine",
         "test_zoomd",
-        "test_vltransformer",
+        "test_transchex",
     ]
     assert sorted(exclude_cases) == sorted(set(exclude_cases)), f"Duplicated items in {exclude_cases}"
 

tests/test_vltransformer.py → tests/test_transchex.py

@@ -15,9 +15,9 @@
 from parameterized import parameterized
 
 from monai.networks import eval_mode
-from monai.networks.nets.vltransformer import VLTransformers
+from monai.networks.nets.transchex import Transchex
 
-TEST_CASE_VLTransformers = []
+TEST_CASE_TRANSCHEX = []
 for drop_out in [0.4]:
     for in_channels in [3]:
         for img_size in [224]:
@@ -39,20 +39,20 @@
                                     },
                                     (2, num_classes),  # type: ignore
                                 ]
-                                TEST_CASE_VLTransformers.append(test_case)
+                                TEST_CASE_TRANSCHEX.append(test_case)
 
 
 class TestPatchEmbeddingBlock(unittest.TestCase):
-    @parameterized.expand(TEST_CASE_VLTransformers)
+    @parameterized.expand(TEST_CASE_TRANSCHEX)
     def test_shape(self, input_param, expected_shape):
-        net = VLTransformers(**input_param)
+        net = Transchex(**input_param)
         with eval_mode(net):
             result = net(torch.randint(2, (2, 512)), torch.randint(2, (2, 512)), torch.randn((2, 3, 224, 224)))
             self.assertEqual(result.shape, expected_shape)
 
     def test_ill_arg(self):
         with self.assertRaises(ValueError):
-            VLTransformers(
+            Transchex(
                 in_channels=3,
                 img_size=(128, 128),
                 patch_size=(16, 16),
@@ -64,7 +64,7 @@ def test_ill_arg(self):
             )
 
         with self.assertRaises(ValueError):
-            VLTransformers(
+            Transchex(
                 in_channels=1,
                 img_size=(97, 97),
                 patch_size=(16, 16),