[Backend] Support onnxruntime DirectML inference. (#1304)

* Fix links in readme * Fix links in readme * Update PPOCRv2/v3 examples * Update auto compression configs * Add neww quantization support for paddleclas model * Update quantized Yolov6s model download link * Improve PPOCR comments * Add English doc for quantization * Fix PPOCR rec model bug * Add new paddleseg quantization support * Add new paddleseg quantization support * Add new paddleseg quantization support * Add new paddleseg quantization support * Add Ascend model list * Add ascend model list * Add ascend model list * Add ascend model list * Add ascend model list * Add ascend model list * Add ascend model list * Support DirectML in onnxruntime * Support onnxruntime DirectML * Support onnxruntime DirectML * Support onnxruntime DirectML * Support OnnxRuntime DirectML * Support OnnxRuntime DirectML * Support OnnxRuntime DirectML * Support OnnxRuntime DirectML * Support OnnxRuntime DirectML * Support OnnxRuntime DirectML * Support OnnxRuntime DirectML * Support OnnxRuntime DirectML * Remove DirectML vision model example * Imporve OnnxRuntime DirectML * Imporve OnnxRuntime DirectML * fix opencv cmake in Windows * recheck codestyle
2025-10-05 16:48:03 +08:00 · 2023-02-17 10:53:51 +08:00
parent efa46563f3
commit c38b7d4377
22 changed files with 393 additions and 60 deletions
--- a/examples/runtime/cpp/infer_paddle_dml.cc
+++ b/examples/runtime/cpp/infer_paddle_dml.cc
@@ -0,0 +1,77 @@
+// Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+#include "fastdeploy/runtime.h"
+
+namespace fd = fastdeploy;
+
+int main(int argc, char* argv[]) {
+  // create option
+  fd::RuntimeOption runtime_option;
+
+  // model and param files
+  std::string model_file = "mobilenetv2/inference.pdmodel";
+  std::string params_file = "mobilenetv2/inference.pdiparams";
+
+  // read model From disk.
+  // runtime_option.SetModelPath(model_file, params_file,
+  // fd::ModelFormat::PADDLE);
+
+  // read model from buffer
+  std::string model_buffer, params_buffer;
+  fd::ReadBinaryFromFile(model_file, &model_buffer);
+  fd::ReadBinaryFromFile(params_file, &params_buffer);
+  runtime_option.SetModelBuffer(model_buffer, params_buffer,
+                                fd::ModelFormat::PADDLE);
+
+  // setup other option
+  runtime_option.SetCpuThreadNum(12);
+  // use ONNX Runtime DirectML
+  runtime_option.UseOrtBackend();
+  runtime_option.UseDirectML();
+
+  // init runtime
+  std::unique_ptr<fd::Runtime> runtime =
+      std::unique_ptr<fd::Runtime>(new fd::Runtime());
+  if (!runtime->Init(runtime_option)) {
+    std::cerr << "--- Init FastDeploy Runitme Failed! "
+              << "\n--- Model:  " << model_file << std::endl;
+    return -1;
+  } else {
+    std::cout << "--- Init FastDeploy Runitme Done! "
+              << "\n--- Model:  " << model_file << std::endl;
+  }
+  // init input tensor shape
+  fd::TensorInfo info = runtime->GetInputInfo(0);
+  info.shape = {1, 3, 224, 224};
+
+  std::vector<fd::FDTensor> input_tensors(1);
+  std::vector<fd::FDTensor> output_tensors(1);
+
+  std::vector<float> inputs_data;
+  inputs_data.resize(1 * 3 * 224 * 224);
+  for (size_t i = 0; i < inputs_data.size(); ++i) {
+    inputs_data[i] = std::rand() % 1000 / 1000.0f;
+  }
+  input_tensors[0].SetExternalData({1, 3, 224, 224}, fd::FDDataType::FP32,
+                                   inputs_data.data());
+
+  // get input name
+  input_tensors[0].name = info.name;
+
+  runtime->Infer(input_tensors, &output_tensors);
+
+  output_tensors[0].PrintInfo();
+  return 0;
+}