fix onnx models convert to quantized bug

mindspore/lite/tools/converter/parser/onnx/onnx_model_parser.cc

@@ -158,16 +158,16 @@ STATUS OnnxModelParser::ConvertNodes() {
       status = RET_ERROR;
       continue;
     }
-    if (IsSpecialOnnxNode(onnx_node)) {
-      auto status_node = ConvertSpecialOnnxNode(onnx_node, primitive_c);
-      status = status == RET_OK ? status_node : status;
-      continue;
-    }
     status = ConvertOpQuantParams(onnx_node, primitive_c);
     if (status != RET_OK) {
       MS_LOG(ERROR) << "convert " << onnx_node.op_type() << " quant param failed.";
       continue;
     }
+    if (IsSpecialOnnxNode(onnx_node)) {
+      auto status_node = ConvertSpecialOnnxNode(onnx_node, primitive_c);
+      status = status == RET_OK ? status_node : status;
+      continue;
+    }
     // build CNode
     status = BuildCNode(onnx_node, primitive_c);
     if (status != RET_OK) {
@@ -512,8 +512,10 @@ STATUS OnnxModelParser::BuildCNodeForGemm(const onnx::NodeProto &onnx_node, lite
         return RET_ERROR;
       } else {
         op_inputs.push_back(nodes_[onnx_node.input(i)]);
+        prim_ptr->AddInputQuantParam(primitive_c->input_quant_params().at(i));
       }
     }
+    prim_ptr->AddOutputQuantParam(std::vector<schema::QuantParamT>(1));
     auto new_cnode = func_graph_ptr_->NewCNode(prim_ptr, op_inputs);
     new_cnode->set_fullname_with_scope("Gemm_MatMul_" + onnx_node.output(0));
     new_cnode->set_abstract(std::make_shared<abstract::AbstractTensor>(type_ptr, shape_vector));
@@ -526,6 +528,9 @@ STATUS OnnxModelParser::BuildCNodeForGemm(const onnx::NodeProto &onnx_node, lite
     }
     op_inputs.push_back(nodes_["Gemm_MatMul_" + onnx_node.output(0)]);
     op_inputs.push_back(nodes_[onnx_node.input(2)]);
+    prim_ptr->AddInputQuantParam(std::vector<schema::QuantParamT>(1));
+    prim_ptr->AddInputQuantParam(primitive_c->input_quant_params().at(2));
+    prim_ptr->AddOutputQuantParam(primitive_c->output_quant_params().front());
     auto new_cnode = func_graph_ptr_->NewCNode(prim_ptr, op_inputs);
     new_cnode->set_fullname_with_scope("Gemm_BiasAdd_" + onnx_node.output(0));
     new_cnode->set_abstract(std::make_shared<abstract::AbstractTensor>(type_ptr, shape_vector));