Removes reliance on engine Keras module, in preparation for Keras 2.2.1. Maintains compatibility with Keras 2.2.0.

Author: fchollet

keras_applications/__init__.py

@@ -11,7 +11,12 @@
 _KERAS_UTILS = None
 
 
-def set_keras_submodules(backend, engine, layers, models, utils):
+def set_keras_submodules(backend=None,
+                         engine=None,
+                         layers=None,
+                         models=None,
+                         utils=None):
+    # TODO: remove `engine` argument after release of Keras 2.2.1.
     global _KERAS_BACKEND
     global _KERAS_ENGINE
     global _KERAS_LAYERS

keras_applications/densenet.py

@@ -231,7 +231,11 @@ def DenseNet(blocks,
     # Ensure that the model takes into account
     # any potential predecessors of `input_tensor`.
     if input_tensor is not None:
-        inputs = engine.get_source_inputs(input_tensor)
+        if hasattr(keras_utils, 'get_source_inputs'):
+            get_source_inputs = keras_utils.get_source_inputs
+        else:
+            get_source_inputs = engine.get_source_inputs
+        inputs = get_source_inputs(input_tensor)
     else:
         inputs = img_input
 

keras_applications/inception_resnet_v2.py

@@ -335,7 +335,11 @@ def InceptionResNetV2(include_top=True,
     # Ensure that the model takes into account
     # any potential predecessors of `input_tensor`.
     if input_tensor is not None:
-        inputs = engine.get_source_inputs(input_tensor)
+        if hasattr(keras_utils, 'get_source_inputs'):
+            get_source_inputs = keras_utils.get_source_inputs
+        else:
+            get_source_inputs = engine.get_source_inputs
+        inputs = get_source_inputs(input_tensor)
     else:
         inputs = img_input
 

keras_applications/inception_v3.py

@@ -366,7 +366,11 @@ def InceptionV3(include_top=True,
     # Ensure that the model takes into account
     # any potential predecessors of `input_tensor`.
     if input_tensor is not None:
-        inputs = engine.get_source_inputs(input_tensor)
+        if hasattr(keras_utils, 'get_source_inputs'):
+            get_source_inputs = keras_utils.get_source_inputs
+        else:
+            get_source_inputs = engine.get_source_inputs
+        inputs = get_source_inputs(input_tensor)
     else:
         inputs = img_input
     # Create model.

keras_applications/mobilenet.py

@@ -282,7 +282,11 @@ def MobileNet(input_shape=None,
     # Ensure that the model takes into account
     # any potential predecessors of `input_tensor`.
     if input_tensor is not None:
-        inputs = engine.get_source_inputs(input_tensor)
+        if hasattr(keras_utils, 'get_source_inputs'):
+            get_source_inputs = keras_utils.get_source_inputs
+        else:
+            get_source_inputs = engine.get_source_inputs
+        inputs = get_source_inputs(input_tensor)
     else:
         inputs = img_input
 

keras_applications/mobilenet_v2.py

@@ -92,6 +92,11 @@
 from .imagenet_utils import decode_predictions
 from .imagenet_utils import _obtain_input_shape
 
+if hasattr(keras_utils, 'get_source_inputs'):
+    get_source_inputs = keras_utils.get_source_inputs
+else:
+    get_source_inputs = engine.get_source_inputs
+
 # TODO Change path to v1.1
 BASE_WEIGHT_PATH = ('https://github.com/JonathanCMitchell/mobilenet_v2_keras/'
                     'releases/download/v1.1/')
@@ -213,7 +218,7 @@ def MobileNetV2(input_shape=None,
         except ValueError:
             try:
                 is_input_t_tensor = backend.is_keras_tensor(
-                    engine.get_source_inputs(input_tensor))
+                    get_source_inputs(input_tensor))
             except ValueError:
                 raise ValueError('input_tensor: ', input_tensor,
                                  'is not type input_tensor')
@@ -417,7 +422,7 @@ def MobileNetV2(input_shape=None,
     # Ensure that the model takes into account
     # any potential predecessors of `input_tensor`.
     if input_tensor is not None:
-        inputs = engine.get_source_inputs(input_tensor)
+        inputs = get_source_inputs(input_tensor)
     else:
         inputs = img_input
 

keras_applications/nasnet.py

@@ -247,7 +247,11 @@ def NASNet(input_shape=None,
     # Ensure that the model takes into account
     # any potential predecessors of `input_tensor`.
     if input_tensor is not None:
-        inputs = engine.get_source_inputs(input_tensor)
+        if hasattr(keras_utils, 'get_source_inputs'):
+            get_source_inputs = keras_utils.get_source_inputs
+        else:
+            get_source_inputs = engine.get_source_inputs
+        inputs = get_source_inputs(input_tensor)
     else:
         inputs = img_input
 

keras_applications/resnet50.py

@@ -266,7 +266,11 @@ def ResNet50(include_top=True,
     # Ensure that the model takes into account
     # any potential predecessors of `input_tensor`.
     if input_tensor is not None:
-        inputs = engine.get_source_inputs(input_tensor)
+        if hasattr(keras_utils, 'get_source_inputs'):
+            get_source_inputs = keras_utils.get_source_inputs
+        else:
+            get_source_inputs = engine.get_source_inputs
+        inputs = get_source_inputs(input_tensor)
     else:
         inputs = img_input
     # Create model.

keras_applications/vgg16.py

@@ -191,7 +191,11 @@ def VGG16(include_top=True,
     # Ensure that the model takes into account
     # any potential predecessors of `input_tensor`.
     if input_tensor is not None:
-        inputs = engine.get_source_inputs(input_tensor)
+        if hasattr(keras_utils, 'get_source_inputs'):
+            get_source_inputs = keras_utils.get_source_inputs
+        else:
+            get_source_inputs = engine.get_source_inputs
+        inputs = get_source_inputs(input_tensor)
     else:
         inputs = img_input
     # Create model.

keras_applications/vgg19.py

@@ -203,7 +203,11 @@ def VGG19(include_top=True,
     # Ensure that the model takes into account
     # any potential predecessors of `input_tensor`.
     if input_tensor is not None:
-        inputs = engine.get_source_inputs(input_tensor)
+        if hasattr(keras_utils, 'get_source_inputs'):
+            get_source_inputs = keras_utils.get_source_inputs
+        else:
+            get_source_inputs = engine.get_source_inputs
+        inputs = get_source_inputs(input_tensor)
     else:
         inputs = img_input
     # Create model.