change labe_list to label_list

deploy/configs/inference_cartoon.yaml

@@ -8,7 +8,7 @@ Global:
   image_shape: [3, 640, 640]
   threshold: 0.2
   max_det_results: 5
-  labe_list:
+  label_list:
   - foreground
 
   use_gpu: True

deploy/configs/inference_det.yaml

@@ -5,7 +5,7 @@ Global:
   image_shape: [3, 640, 640]
   threshold: 0.2
   max_det_results: 1
-  labe_list:
+  label_list:
   - foreground
 
   # inference engine config

deploy/configs/inference_drink.yaml

@@ -8,7 +8,7 @@ Global:
   image_shape: [3, 640, 640]
   threshold: 0.2
   max_det_results: 5
-  labe_list:
+  label_list:
   - foreground
 
   use_gpu: True

deploy/configs/inference_general.yaml

@@ -8,7 +8,7 @@ Global:
   image_shape: [3, 640, 640]
   threshold: 0.2
   max_det_results: 5
-  labe_list:
+  label_list:
   - foreground
 
   use_gpu: True

deploy/configs/inference_general_binary.yaml

@@ -8,7 +8,7 @@ Global:
   image_shape: [3, 640, 640]
   threshold: 0.2
   max_det_results: 5
-  labe_list:
+  label_list:
   - foreground
 
   use_gpu: True

deploy/configs/inference_logo.yaml

@@ -8,7 +8,7 @@ Global:
   image_shape: [3, 640, 640]
   threshold: 0.2
   max_det_results: 5
-  labe_list:
+  label_list:
   - foreground
 
   use_gpu: True

deploy/configs/inference_product.yaml

@@ -8,7 +8,7 @@ Global:
   image_shape: [3, 640, 640]
   threshold: 0.2
   max_det_results: 5
-  labe_list:
+  label_list:
   - foreground
 
   # inference engine config

deploy/configs/inference_vehicle.yaml

@@ -8,7 +8,7 @@ Global:
   image_shape: [3, 640, 640]
   threshold: 0.2
   max_det_results: 5
-  labe_list:
+  label_list:
   - foreground
 
   use_gpu: True

deploy/cpp_shitu/include/object_detector.h

@@ -33,106 +33,106 @@ using namespace paddle_infer;
 
 namespace Detection {
 // Object Detection Result
-    struct ObjectResult {
-        // Rectangle coordinates of detected object: left, right, top, down
-        std::vector<int> rect;
-        // Class id of detected object
-        int class_id;
-        // Confidence of detected object
-        float confidence;
-    };
+struct ObjectResult {
+  // Rectangle coordinates of detected object: left, right, top, down
+  std::vector<int> rect;
+  // Class id of detected object
+  int class_id;
+  // Confidence of detected object
+  float confidence;
+};
 
 // Generate visualization colormap for each class
-    std::vector<int> GenerateColorMap(int num_class);
+std::vector<int> GenerateColorMap(int num_class);
 
 // Visualiztion Detection Result
-    cv::Mat VisualizeResult(const cv::Mat &img,
-                            const std::vector <ObjectResult> &results,
-                            const std::vector <std::string> &lables,
-                            const std::vector<int> &colormap, const bool is_rbox);
-
-    class ObjectDetector {
-    public:
-        explicit ObjectDetector(const YAML::Node &config_file) {
-            this->use_gpu_ = config_file["Global"]["use_gpu"].as<bool>();
-            if (config_file["Global"]["gpu_id"].IsDefined())
-                this->gpu_id_ = config_file["Global"]["gpu_id"].as<int>();
-            this->gpu_mem_ = config_file["Global"]["gpu_mem"].as<int>();
-            this->cpu_math_library_num_threads_ =
-                    config_file["Global"]["cpu_num_threads"].as<int>();
-            this->use_mkldnn_ = config_file["Global"]["enable_mkldnn"].as<bool>();
-            this->use_tensorrt_ = config_file["Global"]["use_tensorrt"].as<bool>();
-            this->use_fp16_ = config_file["Global"]["use_fp16"].as<bool>();
-            this->model_dir_ =
-                    config_file["Global"]["det_inference_model_dir"].as<std::string>();
-            this->threshold_ = config_file["Global"]["threshold"].as<float>();
-            this->max_det_results_ = config_file["Global"]["max_det_results"].as<int>();
-            this->image_shape_ =
-                    config_file["Global"]["image_shape"].as < std::vector < int >> ();
-            this->label_list_ =
-                    config_file["Global"]["labe_list"].as < std::vector < std::string >> ();
-            this->ir_optim_ = config_file["Global"]["ir_optim"].as<bool>();
-            this->batch_size_ = config_file["Global"]["batch_size"].as<int>();
-
-            preprocessor_.Init(config_file["DetPreProcess"]["transform_ops"]);
-            LoadModel(model_dir_, batch_size_, run_mode);
-        }
-
-        // Load Paddle inference model
-        void LoadModel(const std::string &model_dir, const int batch_size = 1,
-                       const std::string &run_mode = "fluid");
-
-        // Run predictor
-        void Predict(const std::vector <cv::Mat> imgs, const int warmup = 0,
-                     const int repeats = 1,
-                     std::vector <ObjectResult> *result = nullptr,
-                     std::vector<int> *bbox_num = nullptr,
-                     std::vector<double> *times = nullptr);
-
-        const std::vector <std::string> &GetLabelList() const {
-            return this->label_list_;
-        }
-
-        const float &GetThreshold() const { return this->threshold_; }
-
-    private:
-        bool use_gpu_ = true;
-        int gpu_id_ = 0;
-        int gpu_mem_ = 800;
-        int cpu_math_library_num_threads_ = 6;
-        std::string run_mode = "fluid";
-        bool use_mkldnn_ = false;
-        bool use_tensorrt_ = false;
-        bool batch_size_ = 1;
-        bool use_fp16_ = false;
-        std::string model_dir_;
-        float threshold_ = 0.5;
-        float max_det_results_ = 5;
-        std::vector<int> image_shape_ = {3, 640, 640};
-        std::vector <std::string> label_list_;
-        bool ir_optim_ = true;
-        bool det_permute_ = true;
-        bool det_postprocess_ = true;
-        int min_subgraph_size_ = 30;
-        bool use_dynamic_shape_ = false;
-        int trt_min_shape_ = 1;
-        int trt_max_shape_ = 1280;
-        int trt_opt_shape_ = 640;
-        bool trt_calib_mode_ = false;
-
-        // Preprocess image and copy data to input buffer
-        void Preprocess(const cv::Mat &image_mat);
-
-        // Postprocess result
-        void Postprocess(const std::vector <cv::Mat> mats,
-                         std::vector <ObjectResult> *result, std::vector<int> bbox_num,
-                         bool is_rbox);
-
-        std::shared_ptr <Predictor> predictor_;
-        Preprocessor preprocessor_;
-        ImageBlob inputs_;
-        std::vector<float> output_data_;
-        std::vector<int> out_bbox_num_data_;
-    };
+cv::Mat VisualizeResult(const cv::Mat &img,
+                        const std::vector<ObjectResult> &results,
+                        const std::vector<std::string> &lables,
+                        const std::vector<int> &colormap, const bool is_rbox);
+
+class ObjectDetector {
+public:
+  explicit ObjectDetector(const YAML::Node &config_file) {
+    this->use_gpu_ = config_file["Global"]["use_gpu"].as<bool>();
+    if (config_file["Global"]["gpu_id"].IsDefined())
+      this->gpu_id_ = config_file["Global"]["gpu_id"].as<int>();
+    this->gpu_mem_ = config_file["Global"]["gpu_mem"].as<int>();
+    this->cpu_math_library_num_threads_ =
+        config_file["Global"]["cpu_num_threads"].as<int>();
+    this->use_mkldnn_ = config_file["Global"]["enable_mkldnn"].as<bool>();
+    this->use_tensorrt_ = config_file["Global"]["use_tensorrt"].as<bool>();
+    this->use_fp16_ = config_file["Global"]["use_fp16"].as<bool>();
+    this->model_dir_ =
+        config_file["Global"]["det_inference_model_dir"].as<std::string>();
+    this->threshold_ = config_file["Global"]["threshold"].as<float>();
+    this->max_det_results_ = config_file["Global"]["max_det_results"].as<int>();
+    this->image_shape_ =
+        config_file["Global"]["image_shape"].as<std::vector<int>>();
+    this->label_list_ =
+        config_file["Global"]["label_list"].as<std::vector<std::string>>();
+    this->ir_optim_ = config_file["Global"]["ir_optim"].as<bool>();
+    this->batch_size_ = config_file["Global"]["batch_size"].as<int>();
+
+    preprocessor_.Init(config_file["DetPreProcess"]["transform_ops"]);
+    LoadModel(model_dir_, batch_size_, run_mode);
+  }
+
+  // Load Paddle inference model
+  void LoadModel(const std::string &model_dir, const int batch_size = 1,
+                 const std::string &run_mode = "fluid");
+
+  // Run predictor
+  void Predict(const std::vector<cv::Mat> imgs, const int warmup = 0,
+               const int repeats = 1,
+               std::vector<ObjectResult> *result = nullptr,
+               std::vector<int> *bbox_num = nullptr,
+               std::vector<double> *times = nullptr);
+
+  const std::vector<std::string> &GetLabelList() const {
+    return this->label_list_;
+  }
+
+  const float &GetThreshold() const { return this->threshold_; }
+
+private:
+  bool use_gpu_ = true;
+  int gpu_id_ = 0;
+  int gpu_mem_ = 800;
+  int cpu_math_library_num_threads_ = 6;
+  std::string run_mode = "fluid";
+  bool use_mkldnn_ = false;
+  bool use_tensorrt_ = false;
+  bool batch_size_ = 1;
+  bool use_fp16_ = false;
+  std::string model_dir_;
+  float threshold_ = 0.5;
+  float max_det_results_ = 5;
+  std::vector<int> image_shape_ = {3, 640, 640};
+  std::vector<std::string> label_list_;
+  bool ir_optim_ = true;
+  bool det_permute_ = true;
+  bool det_postprocess_ = true;
+  int min_subgraph_size_ = 30;
+  bool use_dynamic_shape_ = false;
+  int trt_min_shape_ = 1;
+  int trt_max_shape_ = 1280;
+  int trt_opt_shape_ = 640;
+  bool trt_calib_mode_ = false;
+
+  // Preprocess image and copy data to input buffer
+  void Preprocess(const cv::Mat &image_mat);
+
+  // Postprocess result
+  void Postprocess(const std::vector<cv::Mat> mats,
+                   std::vector<ObjectResult> *result, std::vector<int> bbox_num,
+                   bool is_rbox);
+
+  std::shared_ptr<Predictor> predictor_;
+  Preprocessor preprocessor_;
+  ImageBlob inputs_;
+  std::vector<float> output_data_;
+  std::vector<int> out_bbox_num_data_;
+};
 
 } // namespace Detection

deploy/lite_shitu/generate_json_config.py

@@ -95,7 +95,7 @@ def main():
     config_json["Global"]["det_model_path"] = args.det_model_path
     config_json["Global"]["rec_model_path"] = args.rec_model_path
     config_json["Global"]["rec_label_path"] = args.rec_label_path
-    config_json["Global"]["label_list"] = config_yaml["Global"]["labe_list"]
+    config_json["Global"]["label_list"] = config_yaml["Global"]["label_list"]
     config_json["Global"]["rec_nms_thresold"] = config_yaml["Global"][
         "rec_nms_thresold"]
     config_json["Global"]["max_det_results"] = config_yaml["Global"][

deploy/python/predict_det.py

@@ -134,7 +134,7 @@ class DetPredictor(Predictor):
 
         results = self.parse_det_results(results,
                                          self.config["Global"]["threshold"],
-                                         self.config["Global"]["labe_list"])
+                                         self.config["Global"]["label_list"])
         return results