added flag to add activation on top

model.py

@@ -222,7 +222,7 @@ def _inverted_res_block(inputs, expansion, stride, alpha, filters, block_id, ski
     return x
 
 
-def Deeplabv3(weights='pascal_voc', input_tensor=None, input_shape=(512, 512, 3), classes=21, backbone='mobilenetv2', OS=16, alpha=1.):
+def Deeplabv3(weights='pascal_voc', input_tensor=None, input_shape=(512, 512, 3), classes=21, backbone='mobilenetv2', OS=16, alpha=1., activation=None):
     """ Instantiates the Deeplabv3+ architecture
 
     Optionally loads weights pre-trained
@@ -239,6 +239,8 @@ def Deeplabv3(weights='pascal_voc', input_tensor=None, input_shape=(512, 512, 3)
         classes: number of desired classes. If classes != 21,
             last layer is initialized randomly
         backbone: backbone to use. one of {'xception','mobilenetv2'}
+        activation: optional activation to add to the top of the network. 
+            One of 'softmax', 'sigmoid' or None
         OS: determines input_shape/feature_extractor_output ratio. One of {8,16}.
             Used only for xception backbone.
         alpha: controls the width of the MobileNetV2 network. This is known as the
@@ -249,7 +251,7 @@ def Deeplabv3(weights='pascal_voc', input_tensor=None, input_shape=(512, 512, 3)
                     of filters in each layer.
                 - If `alpha` = 1, default number of filters from the paper
                     are used at each layer.
-            Used only for mobilenetv2 backbone
+            Used only for mobilenetv2 backbone. Pretrained is only available for alpha=1.
 
     # Returns
         A Keras model instance.
@@ -451,6 +453,9 @@ def Deeplabv3(weights='pascal_voc', input_tensor=None, input_shape=(512, 512, 3)
     else:
         inputs = img_input
 
+    if activation in {'softmax','sigmoid'}:
+        x = tf.keras.layers.Activation(activation)(x)
+
     model = Model(inputs, x, name='deeplabv3plus')
 
     # load weights