[ONNX][ONNX Editor] Generate an input with unknown shape and type for…

… operators without schema. (#5531)

ngraph/frontend/onnx_editor/src/detail/subgraph_extraction.cpp

@@ -105,21 +105,31 @@ namespace
 
     /// \brief Looks up a descriptor for a given tensor name. This descriptor contains inferred
     ///        shape information which is required to create new inputs and outputs in the graph.
-    const ONNX_NAMESPACE::ValueInfoProto&
+    const ONNX_NAMESPACE::ValueInfoProto
         find_tensor_descriptor(const ONNX_NAMESPACE::GraphProto& graph,
                                const std::string& tensor_name)
     {
         const auto it = std::find_if(std::begin(graph.value_info()),
                                      std::end(graph.value_info()),
                                      name_equals<ONNX_NAMESPACE::ValueInfoProto>(tensor_name));
 
-        NGRAPH_CHECK(it != std::end(graph.value_info()),
-                     "Could not find a tensor descriptor for tensor '",
-                     tensor_name,
-                     "'. It's not possible to add a new input to the graph without the type and "
-                     "shape information of the intermediate tensor.");
-
-        return *it;
+        if (it != std::end(graph.value_info()))
+        {
+            return *it;
+        }
+        else
+        {
+            // If tensor descriptor couldn't be found value info has to be specified as fully dynamic:
+            // - Fully dynamic shape
+            // - Unknown data type
+            auto dynamic_value_info = ONNX_NAMESPACE::ValueInfoProto();
+            dynamic_value_info.set_name(tensor_name);
+            auto type = dynamic_value_info.mutable_type();
+            auto tensor_type = type->mutable_tensor_type();
+            tensor_type->set_elem_type(
+                ONNX_NAMESPACE::TensorProto_DataType::TensorProto_DataType_UNDEFINED);
+            return dynamic_value_info;
+        }
     }
 
     std::string get_input_tensor_name(const ONNX_NAMESPACE::GraphProto& graph,

ngraph/test/models/onnx/model_editor/reference/unknown_input_value_info.prototxt

@@ -0,0 +1,73 @@
+ir_version: 7
+producer_name: "test_data_generator"
+graph {
+  name: "No name"
+  node {
+    input: "X"
+    output: "X2"
+    name: ""
+    op_type: "Relu"
+  }
+  node {
+    input: "X2"
+    output: "Y"
+    name: ""
+    op_type: "MaxPool"
+    attribute {
+      name: "strides"
+      ints: 2
+      ints: 2
+      type: INTS
+    }
+    attribute {
+      name: "pads"
+      ints: 0
+      ints: 0
+      ints: 0
+      ints: 0
+      type: INTS
+    }
+    attribute {
+      name: "kernel_shape"
+      ints: 3
+      ints: 3
+      type: INTS
+    }
+  }
+
+  input {
+    name: "X"
+    type {
+      tensor_type {
+        elem_type: 0
+      }
+    }
+  }
+
+  output {
+    name: "Y"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim {
+            dim_value: 1
+          }
+          dim {
+            dim_value: 64
+          }
+          dim {
+            dim_value: 55
+          }
+          dim {
+            dim_value: 55
+          }
+        }
+      }
+    }
+  }
+}
+
+opset_import {
+  version: 13
+}

ngraph/test/models/onnx/model_editor/unknown_input_value_info.prototxt

@@ -0,0 +1,132 @@
+ir_version: 7
+producer_name: "test_data_generator"
+graph {
+  name: "No_name"
+  node {
+    input: "A"
+    input: "B"
+    input: "C"
+    output: "X"
+    name: ""
+    op_type: "Bongo"
+  }
+  node {
+    input: "X"
+    output: "X2"
+    name: ""
+    op_type: "Relu"
+  }
+  node {
+    input: "X2"
+    output: "Y"
+    name: ""
+    op_type: "MaxPool"
+    attribute {
+      name: "strides"
+      ints: 2
+      ints: 2
+      type: INTS
+    }
+    attribute {
+      name: "pads"
+      ints: 0
+      ints: 0
+      ints: 0
+      ints: 0
+      type: INTS
+    }
+    attribute {
+      name: "kernel_shape"
+      ints: 3
+      ints: 3
+      type: INTS
+    }
+  }
+
+  input {
+    name: "A"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim {
+            dim_value: 1
+          }
+          dim {
+            dim_value: 3
+          }
+          dim {
+            dim_value: 224
+          }
+          dim {
+            dim_value: 224
+          }
+        }
+      }
+    }
+  }
+
+  input {
+    name: "B"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim {
+            dim_value: 64
+          }
+          dim {
+            dim_value: 3
+          }
+          dim {
+            dim_value: 7
+          }
+          dim {
+            dim_value: 7
+          }
+        }
+      }
+    }
+  }
+
+  input {
+    name: "C"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim {
+            dim_value: 64
+          }
+        }
+      }
+    }
+  }
+
+  output {
+    name: "Y"
+    type {
+      tensor_type {
+        elem_type: 1
+        shape {
+          dim {
+            dim_value: 1
+          }
+          dim {
+            dim_value: 64
+          }
+          dim {
+            dim_value: 55
+          }
+          dim {
+            dim_value: 55
+          }
+        }
+      }
+    }
+  }
+}
+
+opset_import {
+  version: 13
+}

ngraph/test/onnx/onnx_editor.cpp

@@ -1447,3 +1447,18 @@ NGRAPH_TEST(onnx_editor, combined__cut_and_replace_shape)
     EXPECT_TRUE(
         find_input(graph_inputs, "conv1/7x7_s2_1")->get_partial_shape().same_scheme(new_shape));
 }
+
+NGRAPH_TEST(onnx_editor, cut_operator_with_no_schema)
+{
+    ONNXModelEditor editor{file_util::path_join(
+        SERIALIZED_ZOO, "onnx/model_editor/unknown_input_value_info.prototxt")};
+
+    editor.cut_graph_fragment({{InputEdge(1, "X")}}, {});
+
+    const auto ref_model = file_util::path_join(
+        SERIALIZED_ZOO, "onnx/model_editor/reference/unknown_input_value_info.prototxt");
+
+    const auto result = compare_onnx_models(editor.model_string(), ref_model);
+
+    EXPECT_TRUE(result.is_ok) << result.error_message;
+}