Refactor print_summary.

keras/engine/topology.py

@@ -2324,10 +2324,9 @@ class Container(Layer):
         import yaml
         return yaml.dump(self._updated_config(), **kwargs)
 
-    def summary(self, line_length=100, positions=[.33, .55, .67, 1.]):
+    def summary(self, line_length=None, positions=None):
         from keras.utils.layer_utils import print_summary
-        print_summary(self.layers,
-                      getattr(self, 'container_nodes', None),
+        print_summary(self,
                       line_length=line_length,
                       positions=positions)
 

keras/utils/layer_utils.py

@@ -29,13 +29,11 @@ def layer_from_config(config, custom_objects=None):
                                     printable_module_name='layer')
 
 
-def print_summary(layers, relevant_nodes=None,
-                  line_length=100, positions=None):
+def print_summary(model, line_length=None, positions=None):
     """Prints a summary of a layer.
 
     # Arguments
-        layers: list of layers to print summaries of
-        relevant_nodes: list of relevant nodes
+        model: Keras model instance.
         line_length: total length of printed lines
         positions: relative or absolute positions of log elements in each line.
             If not provided, defaults to `[.33, .55, .67, 1.]`.
@@ -43,11 +41,29 @@ def print_summary(layers, relevant_nodes=None,
     # TODO: don't print connectivity for sequential models
     maybe change API to accept a model instance
     """
-    positions = positions or [.33, .55, .67, 1.]
-    if positions[-1] <= 1:
-        positions = [int(line_length * p) for p in positions]
-    # header names for the different log elements
-    to_display = ['Layer (type)', 'Output Shape', 'Param #', 'Connected to']
+
+    if isinstance(model, Sequential):
+        sequential_like = True
+    else:
+        sequential_like = True
+        for k, v in model.nodes_by_depth.items():
+            if len(v) > 1:
+                sequential_like = False
+
+    if sequential_like:
+        line_length = line_length or 65
+        positions = positions or [.45, .85, 1.]
+        if positions[-1] <= 1:
+            positions = [int(line_length * p) for p in positions]
+        # header names for the different log elements
+        to_display = ['Layer (type)', 'Output Shape', 'Param #']
+    else:
+        line_length = line_length or 100
+        positions = positions or [.33, .55, .67, 1.]
+        if positions[-1] <= 1:
+            positions = [int(line_length * p) for p in positions]
+        # header names for the different log elements
+        to_display = ['Layer (type)', 'Output Shape', 'Param #', 'Connected to']
 
     def print_row(fields, positions):
         line = ''
@@ -64,6 +80,16 @@ def print_summary(layers, relevant_nodes=None,
     print('=' * line_length)
 
     def print_layer_summary(layer):
+        try:
+            output_shape = layer.output_shape
+        except AttributeError:
+            output_shape = 'multiple'
+        name = layer.name
+        cls_name = layer.__class__.__name__
+        fields = [name + ' (' + cls_name + ')', output_shape, layer.count_params()]
+        print_row(fields, positions)
+
+    def print_layer_summary_with_connections(layer):
         """Prints a summary for a single layer.
 
         # Arguments
@@ -99,8 +125,12 @@ def print_summary(layers, relevant_nodes=None,
                 fields = ['', '', '', connections[i]]
                 print_row(fields, positions)
 
+    layers = model.layers
     for i in range(len(layers)):
-        print_layer_summary(layers[i])
+        if sequential_like:
+            print_layer_summary(layers[i])
+        else:
+            print_layer_summary_with_connections(layers[i])
         if i == len(layers) - 1:
             print('=' * line_length)
         else: