change response of names interface from list to tree.

mindinsight/datavisual/data_transform/graph/graph.py

@@ -147,28 +147,6 @@ class Graph:
                 nodes.append(node.to_dict())
         return nodes
 
-    def search_node_names(self, content, offset, limit):
-        """
-        Search node names by content.
-
-        Args:
-            content (Union[str, None]): This content can be the key content of the node to search,
-                                        if None, will get all node names.
-            offset (int): An offset for page. Ex, offset is 0, mean current page is 1.
-            limit (int): An offset for page. Ex, offset is 0, mean current page is 1.
-
-        Returns:
-            list[str], a list of node names.
-        """
-        if content is not None:
-            content = content.lower()
-            catch_names = [name for name in self._normal_node_map if content in name.lower()]
-        else:
-            catch_names = list(self._normal_node_map)
-        catch_names = sorted(catch_names)
-        real_offset = offset * limit
-        return catch_names[real_offset:real_offset+limit]
-
     def search_single_node(self, node_name):
         """
         Search node, and return every layer nodes until this node.

mindinsight/datavisual/data_transform/graph/msgraph.py

@@ -16,6 +16,7 @@
 from mindinsight.datavisual.common.log import logger
 from mindinsight.datavisual.proto_files.mindinsight_anf_ir_pb2 import DataType
 from mindinsight.datavisual.common.enums import PluginNameEnum
+from .node_tree import NodeTree
 from .node import Node
 from .node import NodeTypeEnum
 from .graph import Graph
@@ -205,6 +206,128 @@ class MSGraph(Graph):
 
         return data_type_name
 
+    def get_nodes(self, searched_node_list):
+        """
+        Get node tree by a searched_node_list.
+
+        Args:
+            searched_node_list (list[Node]): A list of nodes that
+                matches the given search pattern.
+
+        Returns:
+            A list of dict including the searched nodes.
+            [{
+                "name": "Default",
+                "type": "name_scope",
+                "nodes": [{
+                    "name": "Default/Conv2D1",
+                    "type": "name_scope",
+                    "nodes": [{
+                        ...
+                    }]
+                }]
+            },
+            {
+                "name": "Gradients",
+                "type": "name_scope",
+                "nodes": [{
+                    "name": "Gradients/Default",
+                    "type": "name_scope",
+                    "nodes": [{
+                        ...
+                }]
+            }]
+        """
+        # save the node in the NodeTree
+        root = NodeTree()
+        for node in searched_node_list:
+            self._build_node_tree(root, node.name, node.type)
+
+        # get the searched nodes in the NodeTree and reorganize them
+        searched_list = []
+        self._traverse_node_tree(root, searched_list)
+
+        return searched_list
+
+    def search_leaf_nodes_by_pattern(self, pattern):
+        """
+        Search leaf node by a given pattern.
+
+        Args:
+            pattern (Union[str, None]): The pattern of the node to search,
+                if None, return all node names.
+
+        Returns:
+            list[Node], a list of nodes.
+        """
+        if pattern is not None:
+            pattern = pattern.lower()
+            searched_nodes = [
+                node for name, node in self._leaf_nodes.items()
+                if pattern in name.lower()
+            ]
+        else:
+            searched_nodes = [node for node in self._leaf_nodes.values()]
+        return searched_nodes
+
+    def search_nodes_by_pattern(self, pattern):
+        """
+        Search node by a given pattern.
+
+        Search node which pattern is the part of the last node. Example: pattern=ops, node1=default/ops,
+        node2=default/ops/weight, so node2 will be ignore and only node1 will be return.
+
+        Args:
+            pattern (Union[str, None]): The pattern of the node to search.
+
+        Returns:
+            list[Node], a list of nodes.
+        """
+        searched_nodes = []
+        if pattern and pattern != '/':
+            pattern = pattern.lower()
+            for name, node in self._normal_node_map.items():
+                name = name.lower()
+                pattern_index = name.rfind(pattern)
+                if pattern_index >= 0 and name.find('/', pattern_index + len(pattern)) == -1:
+                    searched_nodes.append(node)
+        return searched_nodes
+
+    def _build_node_tree(self, root, node_name, node_type):
+        """
+        Build node tree.
+
+        Args:
+            root (NodeTree): Root node of node tree.
+            node_name (str): Node name.
+            node_type (str): Node type.
+        """
+        scope_names = node_name.split('/')
+        cur_node = root
+        full_name = ""
+        for scope_name in scope_names[:-1]:
+            full_name = '/'.join([full_name, scope_name]) if full_name else scope_name
+            scope_node = self._get_normal_node(node_name=full_name)
+            sub_node = cur_node.get(scope_name)
+            if not sub_node:
+                sub_node = cur_node.add(scope_name, scope_node.type)
+            cur_node = sub_node
+        cur_node.add(scope_names[-1], node_type)
+
+    def _traverse_node_tree(self, cur_node, search_node_list):
+        """Traverse the node tree and construct the searched nodes list."""
+        if not cur_node.get_children():
+            return
+        for _, sub_node in cur_node.get_children():
+            sub_nodes = []
+            self._traverse_node_tree(sub_node, sub_nodes)
+            sub_node_dict = {
+                'name': sub_node.node_name,
+                'type': sub_node.node_type,
+                'nodes': sub_nodes
+            }
+            search_node_list.append(sub_node_dict)
+
     def _parse_inputs(self, input_protos, node):
         """
         Parse `anf_ir_pb2.InputProto` object.

mindinsight/datavisual/processors/graph_processor.py

@@ -103,12 +103,15 @@ class GraphProcessor(BaseProcessor):
             limit (int): The max data items for per page.
 
         Returns:
-            TypedDict('Names', {'names': list[str]}), {"names": ["node_names"]}.
+            Dict, the searched nodes.
         """
         offset = Validation.check_offset(offset=offset)
         limit = Validation.check_limit(limit, min_value=1, max_value=1000)
-        names = self._graph.search_node_names(search_content, offset, limit)
-        return {"names": names}
+        nodes = self._graph.search_nodes_by_pattern(search_content)
+        real_offset = offset * limit
+        search_nodes = self._graph.get_nodes(nodes[real_offset:real_offset + limit])
+
+        return {"nodes": search_nodes}
 
     def search_single_node(self, name):
         """

mindinsight/debugger/stream_cache/debugger_graph.py

@@ -19,14 +19,10 @@ from mindinsight.datavisual.data_transform.graph.msgraph import MSGraph
 from mindinsight.debugger.common.exceptions.exceptions import \
     DebuggerNodeNotInGraphError, DebuggerParamValueError
 from mindinsight.debugger.common.log import logger as log
-from .node import NodeTree
 
 
 class DebuggerGraph(MSGraph):
     """The `DebuggerGraph` object provides interfaces to describe a debugger graph."""
-    def __init__(self):
-        super(DebuggerGraph, self).__init__()
-        self._node_tree = None
 
     def get_node_name_by_full_name(self, full_name):
         """Get node name by full names."""
@@ -44,95 +40,6 @@ class DebuggerGraph(MSGraph):
 
         return node.full_name if node else ''
 
-    def get_nodes(self, searched_node_list):
-        """
-        Search node names by a given pattern.
-
-        Args:
-            searched_node_list (list[Node]): A list of leaf nodes that
-                matches the given search pattern.
-
-        Returns:
-            A list of dict including the searched nodes.
-            [{
-                "name": "Default",
-                "type": "name_scope",
-                "nodes": [{
-                    "name": "Default/Conv2D1",
-                    "type": "name_scope",
-                    "nodes": [{
-                        ...
-                    }]
-                }]
-            },
-            {
-                "name": "Gradients",
-                "type": "name_scope",
-                "nodes": [{
-                    "name": "Gradients/Default",
-                    "type": "name_scope",
-                    "nodes": [{
-                        ...
-                }]
-            }]
-        """
-        # save the node in the NodeTree
-        self._node_tree = NodeTree()
-        for node in searched_node_list:
-            self._build_node_tree(node.name, node.type)
-
-        # get the searched nodes in the NodeTree and reorganize them
-        searched_list = []
-        self._traverse_node_tree(self._node_tree, searched_list)
-
-        return searched_list
-
-    def search_nodes_by_pattern(self, pattern):
-        """
-        Search node by a given pattern.
-
-        Args:
-            pattern (Union[str, None]): The pattern of the node to search,
-                if None, return all node names.
-
-        Returns:
-            list[Node], a list of node.
-        """
-        if pattern is not None:
-            pattern = pattern.lower()
-            searched_nodes = [
-                node for name, node in self._leaf_nodes.items()
-                if pattern in name.lower()
-            ]
-        else:
-            searched_nodes = [node for _, node in self._leaf_nodes.items()]
-        return searched_nodes
-
-    def _build_node_tree(self, node_name, node_type):
-        """Build node tree."""
-        scope_names = node_name.split('/')
-        cur_node = self._node_tree
-        for scope_name in scope_names[:-1]:
-            sub_node = cur_node.get(scope_name)
-            if not sub_node:
-                sub_node = cur_node.add(scope_name)
-            cur_node = sub_node
-        cur_node.add(scope_names[-1], node_type)
-
-    def _traverse_node_tree(self, cur_node, search_node_list):
-        """Traverse the watch nodes and update the total watched node list."""
-        if not cur_node.get_children():
-            return
-        for _, sub_node in cur_node.get_children():
-            sub_nodes = []
-            self._traverse_node_tree(sub_node, sub_nodes)
-            sub_node_dict = {
-                'name': sub_node.node_name,
-                'type': sub_node.node_type,
-                'nodes': sub_nodes
-            }
-            search_node_list.append(sub_node_dict)
-
     def get_node_type(self, node_name):
         """
         Get the type of the node.

mindinsight/debugger/stream_handler/graph_handler.py

@@ -153,7 +153,7 @@ class GraphHandler(StreamHandlerBase):
         Returns:
             list[Node], a list of node.
         """
-        return self._graph.search_nodes_by_pattern(scope_name)
+        return self._graph.search_leaf_nodes_by_pattern(scope_name)
 
     def get_searched_node_list(self):
         """Get searched node list."""

tests/st/func/datavisual/graph/graph_results/test_search_nodes_success_result1.json

@@ -1 +1 @@
-{"names":["Default/bn1","Default/bn1-BatchNorm2d","Default/bn1-BatchNorm2d/Parameter[22]_3","Default/bn1-BatchNorm2d/Parameter[22]_3/conv1.weight","Default/bn1-BatchNorm2d/Parameter[22]_3/x","Default/bn1-BatchNorm2d/Parameter[22]_3/x1","Default/bn1-BatchNorm2d/Parameter[22]_3/x10","Default/bn1-BatchNorm2d/Parameter[22]_3/x11","Default/bn1-BatchNorm2d/Parameter[22]_3/x12","Default/bn1-BatchNorm2d/Parameter[22]_3/x13","Default/bn1-BatchNorm2d/Parameter[22]_3/x14","Default/bn1-BatchNorm2d/Parameter[22]_3/x15","Default/bn1-BatchNorm2d/Parameter[22]_3/x16","Default/bn1-BatchNorm2d/Parameter[22]_3/x17","Default/bn1-BatchNorm2d/Parameter[22]_3/x18","Default/bn1-BatchNorm2d/Parameter[22]_3/x19","Default/bn1-BatchNorm2d/Parameter[22]_3/x2","Default/bn1-BatchNorm2d/Parameter[22]_3/x20","Default/bn1-BatchNorm2d/Parameter[22]_3/x3","Default/bn1-BatchNorm2d/Parameter[22]_3/x4","Default/bn1-BatchNorm2d/Parameter[22]_3/x5","Default/bn1-BatchNorm2d/Parameter[22]_3/x6","Default/bn1-BatchNorm2d/Parameter[22]_3/x7","Default/bn1-BatchNorm2d/Parameter[22]_3/x8","Default/bn1-BatchNorm2d/Parameter[22]_3/x9","Default/bn1-BatchNorm2d/cst13","Default/bn1-BatchNorm2d/cst25","Default/bn1-BatchNorm2d/tuple_getitem105","Default/bn1-BatchNorm2d/tuple_getitem56","Default/bn1/Add[5]_0","Default/bn1/Add[5]_0/Add50","Default/bn1/Add[5]_0/Add51","Default/bn1/Add[5]_0/Add52","Default/bn1/Add[5]_0/Add53","Default/bn1/Add[5]_0/Add54","Default/bn1/Reshape[12]_1","Default/bn1/Reshape[12]_1/Reshape1","Default/bn1/Reshape[12]_1/Reshape10","Default/bn1/Reshape[12]_1/Reshape11","Default/bn1/Reshape[12]_1/Reshape12","Default/bn1/Reshape[12]_1/Reshape2","Default/bn1/Reshape[12]_1/Reshape3","Default/bn1/Reshape[12]_1/Reshape4","Default/bn1/Reshape[12]_1/Reshape5","Default/bn1/Reshape[12]_1/Reshape6","Default/bn1/Reshape[12]_1/Reshape7","Default/bn1/Reshape[12]_1/Reshape8","Default/bn1/Reshape[12]_1/Reshape9","Default/bn1/x","Default/bn1/x11"]}
+{"nodes": [{"name": "Default", "type": "name_scope", "nodes": [{"name": "Default/bn1-BatchNorm2d", "type": "name_scope", "nodes": []}, {"name": "Default/bn1", "type": "name_scope", "nodes": []}]}]}

tests/st/func/debugger/expect_results/restful_results/search_unwatched_leaf_node.json

@@ -1 +1 @@
-{"nodes": [{"name": "Default", "type": null, "nodes": [{"name": "Default/optimizer-Momentum", "type": null, "nodes": [{"name": "Default/optimizer-Momentum/Parameter[18]_7", "type": null, "nodes": [{"name": "Default/optimizer-Momentum/Parameter[18]_7/moments.fc3.bias", "type": "Parameter", "nodes": [], "watched": 0}], "watched": 1}], "watched": 1}], "watched": 1}]}
+{"nodes": [{"name": "Default", "type": "name_scope", "nodes": [{"name": "Default/optimizer-Momentum", "type": "name_scope", "nodes": [{"name": "Default/optimizer-Momentum/Parameter[18]_7", "type": "aggregation_scope", "nodes": [{"name": "Default/optimizer-Momentum/Parameter[18]_7/moments.fc3.bias", "type": "Parameter", "nodes": [], "watched": 0}], "watched": 1}], "watched": 1}], "watched": 1}]}

tests/ut/datavisual/processors/graph_results/test_search_node_names_with_offset_expected_results1.json

@@ -1 +1 @@
-{"names":["Default/bn1-BatchNorm2d/Parameter[22]_3/conv1.weight","Default/bn1-BatchNorm2d/Parameter[22]_3/x","Default/bn1-BatchNorm2d/Parameter[22]_3/x1"]}
+{"nodes": [{"name": "Default", "type": "name_scope", "nodes": [{"name": "Default/bn1", "type": "name_scope", "nodes": []}]}]}

tests/ut/datavisual/processors/graph_results/test_search_node_names_with_search_content_expected_results1.json

@@ -1 +1 @@
-{"names":["Default","Default/bn1","Default/bn1-BatchNorm2d","Default/bn1-BatchNorm2d/Parameter[22]_3","Default/bn1-BatchNorm2d/Parameter[22]_3/conv1.weight","Default/bn1-BatchNorm2d/Parameter[22]_3/x","Default/bn1-BatchNorm2d/Parameter[22]_3/x1","Default/bn1-BatchNorm2d/Parameter[22]_3/x10","Default/bn1-BatchNorm2d/Parameter[22]_3/x11","Default/bn1-BatchNorm2d/Parameter[22]_3/x12","Default/bn1-BatchNorm2d/Parameter[22]_3/x13","Default/bn1-BatchNorm2d/Parameter[22]_3/x14","Default/bn1-BatchNorm2d/Parameter[22]_3/x15","Default/bn1-BatchNorm2d/Parameter[22]_3/x16","Default/bn1-BatchNorm2d/Parameter[22]_3/x17","Default/bn1-BatchNorm2d/Parameter[22]_3/x18","Default/bn1-BatchNorm2d/Parameter[22]_3/x19","Default/bn1-BatchNorm2d/Parameter[22]_3/x2","Default/bn1-BatchNorm2d/Parameter[22]_3/x20","Default/bn1-BatchNorm2d/Parameter[22]_3/x3","Default/bn1-BatchNorm2d/Parameter[22]_3/x4","Default/bn1-BatchNorm2d/Parameter[22]_3/x5","Default/bn1-BatchNorm2d/Parameter[22]_3/x6","Default/bn1-BatchNorm2d/Parameter[22]_3/x7","Default/bn1-BatchNorm2d/Parameter[22]_3/x8","Default/bn1-BatchNorm2d/Parameter[22]_3/x9","Default/bn1-BatchNorm2d/cst13","Default/bn1-BatchNorm2d/cst25","Default/bn1-BatchNorm2d/tuple_getitem105","Default/bn1-BatchNorm2d/tuple_getitem56","Default/bn1/Add[5]_0","Default/bn1/Add[5]_0/Add50","Default/bn1/Add[5]_0/Add51","Default/bn1/Add[5]_0/Add52","Default/bn1/Add[5]_0/Add53","Default/bn1/Add[5]_0/Add54","Default/bn1/Reshape[12]_1","Default/bn1/Reshape[12]_1/Reshape1","Default/bn1/Reshape[12]_1/Reshape10","Default/bn1/Reshape[12]_1/Reshape11","Default/bn1/Reshape[12]_1/Reshape12","Default/bn1/Reshape[12]_1/Reshape2","Default/bn1/Reshape[12]_1/Reshape3","Default/bn1/Reshape[12]_1/Reshape4","Default/bn1/Reshape[12]_1/Reshape5","Default/bn1/Reshape[12]_1/Reshape6","Default/bn1/Reshape[12]_1/Reshape7","Default/bn1/Reshape[12]_1/Reshape8","Default/bn1/Reshape[12]_1/Reshape9","Default/bn1/x","Default/bn1/x11","Default/conv1-Conv2d","Default/conv1-Conv2d/Conv2D55","Default/conv1-Conv2d/Parameter[12]_2","Default/conv1-Conv2d/Parameter[12]_2/conv1.weight","Default/conv1-Conv2d/Parameter[12]_2/x","Default/conv1-Conv2d/Parameter[12]_2/x1","Default/conv1-Conv2d/Parameter[12]_2/x10","Default/conv1-Conv2d/Parameter[12]_2/x2","Default/conv1-Conv2d/Parameter[12]_2/x3","Default/conv1-Conv2d/Parameter[12]_2/x4","Default/conv1-Conv2d/Parameter[12]_2/x5","Default/conv1-Conv2d/Parameter[12]_2/x6","Default/conv1-Conv2d/Parameter[12]_2/x7","Default/conv1-Conv2d/Parameter[12]_2/x8","Default/conv1-Conv2d/Parameter[12]_2/x9"]}
+{"nodes": []}

tests/ut/datavisual/processors/graph_results/test_search_node_names_with_search_content_expected_results2.json

@@ -1 +1 @@
-{"names":["Default/bn1","Default/bn1-BatchNorm2d","Default/bn1-BatchNorm2d/Parameter[22]_3","Default/bn1-BatchNorm2d/Parameter[22]_3/conv1.weight","Default/bn1-BatchNorm2d/Parameter[22]_3/x","Default/bn1-BatchNorm2d/Parameter[22]_3/x1","Default/bn1-BatchNorm2d/Parameter[22]_3/x10","Default/bn1-BatchNorm2d/Parameter[22]_3/x11","Default/bn1-BatchNorm2d/Parameter[22]_3/x12","Default/bn1-BatchNorm2d/Parameter[22]_3/x13","Default/bn1-BatchNorm2d/Parameter[22]_3/x14","Default/bn1-BatchNorm2d/Parameter[22]_3/x15","Default/bn1-BatchNorm2d/Parameter[22]_3/x16","Default/bn1-BatchNorm2d/Parameter[22]_3/x17","Default/bn1-BatchNorm2d/Parameter[22]_3/x18","Default/bn1-BatchNorm2d/Parameter[22]_3/x19","Default/bn1-BatchNorm2d/Parameter[22]_3/x2","Default/bn1-BatchNorm2d/Parameter[22]_3/x20","Default/bn1-BatchNorm2d/Parameter[22]_3/x3","Default/bn1-BatchNorm2d/Parameter[22]_3/x4","Default/bn1-BatchNorm2d/Parameter[22]_3/x5","Default/bn1-BatchNorm2d/Parameter[22]_3/x6","Default/bn1-BatchNorm2d/Parameter[22]_3/x7","Default/bn1-BatchNorm2d/Parameter[22]_3/x8","Default/bn1-BatchNorm2d/Parameter[22]_3/x9","Default/bn1-BatchNorm2d/cst13","Default/bn1-BatchNorm2d/cst25","Default/bn1-BatchNorm2d/tuple_getitem105","Default/bn1-BatchNorm2d/tuple_getitem56","Default/bn1/Add[5]_0","Default/bn1/Add[5]_0/Add50","Default/bn1/Add[5]_0/Add51","Default/bn1/Add[5]_0/Add52","Default/bn1/Add[5]_0/Add53","Default/bn1/Add[5]_0/Add54","Default/bn1/Reshape[12]_1","Default/bn1/Reshape[12]_1/Reshape1","Default/bn1/Reshape[12]_1/Reshape10","Default/bn1/Reshape[12]_1/Reshape11","Default/bn1/Reshape[12]_1/Reshape12","Default/bn1/Reshape[12]_1/Reshape2","Default/bn1/Reshape[12]_1/Reshape3","Default/bn1/Reshape[12]_1/Reshape4","Default/bn1/Reshape[12]_1/Reshape5","Default/bn1/Reshape[12]_1/Reshape6","Default/bn1/Reshape[12]_1/Reshape7","Default/bn1/Reshape[12]_1/Reshape8","Default/bn1/Reshape[12]_1/Reshape9","Default/bn1/x","Default/bn1/x11"]}
+{"nodes": [{"name": "Default", "type": "name_scope", "nodes": [{"name": "Default/bn1-BatchNorm2d", "type": "name_scope", "nodes": []}, {"name": "Default/bn1", "type": "name_scope", "nodes": []}]}]}

tests/ut/datavisual/processors/test_graph_processor.py

@@ -149,7 +149,7 @@ class TestGraphProcessor:
         graph_processor = GraphProcessor(self._train_id, self._mock_data_manager)
         results = graph_processor.search_node_names(search_content, test_offset, test_limit)
         if search_content == 'not_exist_search_content':
-            expected_results = {'names': []}
+            expected_results = {'nodes': []}
             assert results == expected_results
         else:
             expected_file_path = os.path.join(self.graph_results_dir, result_file)
@@ -173,7 +173,7 @@ class TestGraphProcessor:
         """Test search node names with offset."""
         test_search_content = "Default/bn1"
         test_offset = offset
-        test_limit = 3
+        test_limit = 1
 
         graph_processor = GraphProcessor(self._train_id, self._mock_data_manager)
         results = graph_processor.search_node_names(test_search_content, test_offset, test_limit)

tests/ut/debugger/expected_results/graph/search_node_1.json

@@ -1 +1 @@
-{"nodes": [{"name": "Default", "type": null, "nodes": [{"name": "Default/network-WithLossCell", "type": null, "nodes": [{"name": "Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits", "type": null, "nodes": [{"name": "Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits/cst1", "type": "Const", "nodes": []}, {"name": "Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits/cst2", "type": "Const", "nodes": []}]}]}, {"name": "Default/tuple_getitem[10]_0", "type": null, "nodes": [{"name": "Default/tuple_getitem[10]_0/cst3", "type": "Const", "nodes": []}, {"name": "Default/tuple_getitem[10]_0/cst4", "type": "Const", "nodes": []}, {"name": "Default/tuple_getitem[10]_0/cst5", "type": "Const", "nodes": []}, {"name": "Default/tuple_getitem[10]_0/cst6", "type": "Const", "nodes": []}, {"name": "Default/tuple_getitem[10]_0/cst7", "type": "Const", "nodes": []}, {"name": "Default/tuple_getitem[10]_0/cst27", "type": "Const", "nodes": []}, {"name": "Default/tuple_getitem[10]_0/cst28", "type": "Const", "nodes": []}, {"name": "Default/tuple_getitem[10]_0/cst32", "type": "Const", "nodes": []}, {"name": "Default/tuple_getitem[10]_0/cst33", "type": "Const", "nodes": []}, {"name": "Default/tuple_getitem[10]_0/cst37", "type": "Const", "nodes": []}]}, {"name": "Default/optimizer-Momentum", "type": null, "nodes": [{"name": "Default/optimizer-Momentum/tuple_getitem[16]_2", "type": null, "nodes": [{"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst9", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst10", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst12", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst13", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst15", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst16", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst18", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst19", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst21", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst22", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst24", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst25", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst29", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst30", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst34", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst35", "type": "Const", "nodes": []}]}, {"name": "Default/optimizer-Momentum/Depend[8]_4", "type": null, "nodes": [{"name": "Default/optimizer-Momentum/Depend[8]_4/cst11", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/Depend[8]_4/cst14", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/Depend[8]_4/cst17", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/Depend[8]_4/cst20", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/Depend[8]_4/cst23", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/Depend[8]_4/cst26", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/Depend[8]_4/cst31", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/Depend[8]_4/cst36", "type": "Const", "nodes": []}]}]}]}, {"name": "Gradients", "type": null, "nodes": [{"name": "Gradients/Default", "type": null, "nodes": [{"name": "Gradients/Default/network-WithLossCell", "type": null, "nodes": [{"name": "Gradients/Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits", "type": null, "nodes": [{"name": "Gradients/Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits/gradSoftmaxCrossEntropyWithLogits", "type": null, "nodes": [{"name": "Gradients/Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits/gradSoftmaxCrossEntropyWithLogits/cst8", "type": "Const", "nodes": []}]}]}]}]}]}]}
+{"nodes": [{"name": "Default", "type": "name_scope", "nodes": [{"name": "Default/network-WithLossCell", "type": "name_scope", "nodes": [{"name": "Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits", "type": "name_scope", "nodes": [{"name": "Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits/cst1", "type": "Const", "nodes": []}, {"name": "Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits/cst2", "type": "Const", "nodes": []}]}]}, {"name": "Default/tuple_getitem[10]_0", "type": "aggregation_scope", "nodes": [{"name": "Default/tuple_getitem[10]_0/cst3", "type": "Const", "nodes": []}, {"name": "Default/tuple_getitem[10]_0/cst4", "type": "Const", "nodes": []}, {"name": "Default/tuple_getitem[10]_0/cst5", "type": "Const", "nodes": []}, {"name": "Default/tuple_getitem[10]_0/cst6", "type": "Const", "nodes": []}, {"name": "Default/tuple_getitem[10]_0/cst7", "type": "Const", "nodes": []}, {"name": "Default/tuple_getitem[10]_0/cst27", "type": "Const", "nodes": []}, {"name": "Default/tuple_getitem[10]_0/cst28", "type": "Const", "nodes": []}, {"name": "Default/tuple_getitem[10]_0/cst32", "type": "Const", "nodes": []}, {"name": "Default/tuple_getitem[10]_0/cst33", "type": "Const", "nodes": []}, {"name": "Default/tuple_getitem[10]_0/cst37", "type": "Const", "nodes": []}]}, {"name": "Default/optimizer-Momentum", "type": "name_scope", "nodes": [{"name": "Default/optimizer-Momentum/tuple_getitem[16]_2", "type": "aggregation_scope", "nodes": [{"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst9", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst10", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst12", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst13", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst15", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst16", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst18", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst19", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst21", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst22", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst24", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst25", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst29", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst30", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst34", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/tuple_getitem[16]_2/cst35", "type": "Const", "nodes": []}]}, {"name": "Default/optimizer-Momentum/Depend[8]_4", "type": "aggregation_scope", "nodes": [{"name": "Default/optimizer-Momentum/Depend[8]_4/cst11", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/Depend[8]_4/cst14", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/Depend[8]_4/cst17", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/Depend[8]_4/cst20", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/Depend[8]_4/cst23", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/Depend[8]_4/cst26", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/Depend[8]_4/cst31", "type": "Const", "nodes": []}, {"name": "Default/optimizer-Momentum/Depend[8]_4/cst36", "type": "Const", "nodes": []}]}]}]}, {"name": "Gradients", "type": "name_scope", "nodes": [{"name": "Gradients/Default", "type": "name_scope", "nodes": [{"name": "Gradients/Default/network-WithLossCell", "type": "name_scope", "nodes": [{"name": "Gradients/Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits", "type": "name_scope", "nodes": [{"name": "Gradients/Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits/gradSoftmaxCrossEntropyWithLogits", "type": "name_scope", "nodes": [{"name": "Gradients/Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits/gradSoftmaxCrossEntropyWithLogits/cst8", "type": "Const", "nodes": []}]}]}]}]}]}]}

tests/ut/debugger/expected_results/graph/search_nodes_0.json

@@ -1 +1 @@
-{"nodes": [{"name": "Default", "type": null, "nodes": [{"name": "Default/network-WithLossCell", "type": null, "nodes": [{"name": "Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits", "type": null, "nodes": [{"name": "Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits/OneHot-op0", "type": "OneHot", "nodes": []}, {"name": "Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits/SoftmaxCrossEntropyWithLogits-op18", "type": "SoftmaxCrossEntropyWithLogits", "nodes": []}, {"name": "Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits/ReduceMean-op60", "type": "ReduceMean", "nodes": []}, {"name": "Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits/args1", "type": "Parameter", "nodes": []}, {"name": "Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits/cst1", "type": "Const", "nodes": []}, {"name": "Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits/cst2", "type": "Const", "nodes": []}]}]}]}, {"name": "Gradients", "type": null, "nodes": [{"name": "Gradients/Default", "type": null, "nodes": [{"name": "Gradients/Default/network-WithLossCell", "type": null, "nodes": [{"name": "Gradients/Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits", "type": null, "nodes": [{"name": "Gradients/Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits/gradSoftmaxCrossEntropyWithLogits", "type": null, "nodes": [{"name": "Gradients/Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits/gradSoftmaxCrossEntropyWithLogits/Mul-op20", "type": "Mul", "nodes": []}, {"name": "Gradients/Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits/gradSoftmaxCrossEntropyWithLogits/cst8", "type": "Const", "nodes": []}]}]}]}]}]}]}
+{"nodes": [{"name": "Default", "type": "name_scope", "nodes": [{"name": "Default/network-WithLossCell", "type": "name_scope", "nodes": [{"name": "Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits", "type": "name_scope", "nodes": []}]}]}, {"name": "Gradients", "type": "name_scope", "nodes": [{"name": "Gradients/Default", "type": "name_scope", "nodes": [{"name": "Gradients/Default/network-WithLossCell", "type": "name_scope", "nodes": [{"name": "Gradients/Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits", "type": "name_scope", "nodes": [{"name": "Gradients/Default/network-WithLossCell/_loss_fn-SoftmaxCrossEntropyWithLogits/gradSoftmaxCrossEntropyWithLogits", "type": "name_scope", "nodes": []}]}]}]}]}]}