TensorFlow Object Detection API中的Faster R-CNN /SSD模子参数调解

关于TensorFlow Object Detection API设置，能够参考之前的文章https://becominghuman.ai/tensorflow-object-detection-api-tutorial-training-and-evaluating-custom-object-detector-ed2594afcf73

在本文中，我将议论怎样变动预练习模子的设置。本文的目的是您能够依据您的应用顺序设置TensorFlow/models，而API将不再是一个黑盒！

本文的概述：

• 相识协定缓冲区和proto文件。
• 应用proto文件学问，我们怎样相识模子的设置文件
• 遵照3个步骤来更新模子的参数
• 其他示例：

1. 变动分量初始值设定项
2. 变动体重优化器
3. 评价预练习模子

协定缓冲区

要修正模子，我们须要相识它的内部机制。TensorFlow对象检测API运用协定缓冲区（Protocol Buffers），这是与言语无关，与平台无关且可扩大的机制，用于序列化结构化数据。就像XML范围较小，但更快，更简朴。API运用协定缓冲区言语的proto2版本。我将尝试诠释更新预设置模子所需的言语。有关协定缓冲区言语的更多详细信息，请参阅此文档和Python教程。

协定缓冲区的事情可分为以下三个步骤：

• 在.proto文件中定义音讯花样。该文件的行动就像一切音讯的蓝图一样，它显现音讯所接收的一切参数是什么，参数的数据范例应当是什么，参数是必须的照样可选的，参数的标暗号是什么，什么是参数的默认值等。API的protos文件可在此处找到。为了明白，我运用grid_anchor_generator.proto文件。

• syntax = "proto2";
  
  package object_detection.protos;
  
  // Configuration proto for GridAnchorGenerator. See
  // anchor_generators/grid_anchor_generator.py for details.
  message GridAnchorGenerator {
     // Anchor height in pixels.
    optional int32 height = 1 [default = 256];
  
    // Anchor width in pixels.
    optional int32 width = 2 [default = 256];
  
    // Anchor stride in height dimension in pixels.
    optional int32 height_stride = 3 [default = 16];
  
    // Anchor stride in width dimension in pixels.
    optional int32 width_stride = 4 [default = 16];
  
    // Anchor height offset in pixels.
    optional int32 height_offset = 5 [default = 0];
  
    // Anchor width offset in pixels.
    optional int32 width_offset = 6 [default = 0];
  
    // At any given location, len(scales) * len(aspect_ratios) anchors are
    // generated with all possible combinations of scales and aspect ratios.
  
    // List of scales for the anchors.
    repeated float scales = 7;
  
    // List of aspect ratios for the anchors.
    repeated float aspect_ratios = 8;
  }

  它是从线30-33的参数明白scales，并aspect_ratios是强制性的音讯GridAnchorGenerator，而参数的其余部份都是可选的，假如不经由过程，将采用默认值。

  • 定义音讯花样后，我们须要编译协定缓冲区。该编译器将从文件生成类.proto文件。在装置API的过程当中，我们运转了以下敕令，该敕令将编译协定缓冲区：

  • # From tensorflow/models/research/
    protoc object_detection/protos/*.proto --python_out=.

    • 在定义和编译协定缓冲区以后，我们须要运用Python协定缓冲区API来写入和读取音讯。在我们的例子中，我们能够将设置文件视为协定缓冲区API，它能够在不斟酌TensorFlow API的内部机制的情况下写入和读取音讯。换句话说，我们能够经由过程适当地变动设置文件来更新预练习模子的参数。

    • 相识设置文件

      显著，设置文件能够协助我们依据须要变动模子的参数。弹出的下一个题目是怎样变动模子的参数？本节和下一部份将回覆这个题目，在这里proto文件的学问将很轻易。出于演示目的，我正在运用faster_rcnn_resnet50_pets.config文件。

    • # Faster R-CNN with Resnet-50 (v1), configured for Oxford-IIIT Pets Dataset.
      # Users should configure the fine_tune_checkpoint field in the train config as
      # well as the label_map_path and input_path fields in the train_input_reader and
      # eval_input_reader. Search for "PATH_TO_BE_CONFIGURED" to find the fields that
      # should be configured.
      
      model {
        faster_rcnn {
          num_classes: 37
          image_resizer {
            keep_aspect_ratio_resizer {
              min_dimension: 600
              max_dimension: 1024
            }
          }
          feature_extractor {
            type: 'faster_rcnn_resnet50'
            first_stage_features_stride: 16
          }
          first_stage_anchor_generator {
            grid_anchor_generator {
              scales: [0.25, 0.5, 1.0, 2.0]
              aspect_ratios: [0.5, 1.0, 2.0]
              height_stride: 16
              width_stride: 16
            }
          }
          first_stage_box_predictor_conv_hyperparams {
            op: CONV
            regularizer {
              l2_regularizer {
                weight: 0.0
              }
            }
            initializer {
              truncated_normal_initializer {
                stddev: 0.01
              }
            }
          }
          first_stage_nms_score_threshold: 0.0
          first_stage_nms_iou_threshold: 0.7
          first_stage_max_proposals: 300
          first_stage_localization_loss_weight: 2.0
          first_stage_objectness_loss_weight: 1.0
          initial_crop_size: 14
          maxpool_kernel_size: 2
          maxpool_stride: 2
          second_stage_box_predictor {
            mask_rcnn_box_predictor {
              use_dropout: false
              dropout_keep_probability: 1.0
              fc_hyperparams {
                op: FC
                regularizer {
                  l2_regularizer {
                    weight: 0.0
                  }
                }
                initializer {
                  variance_scaling_initializer {
                    factor: 1.0
                    uniform: true
                    mode: FAN_AVG
                  }
                }
              }
            }
          }
          second_stage_post_processing {
            batch_non_max_suppression {
              score_threshold: 0.0
              iou_threshold: 0.6
              max_detections_per_class: 100
              max_total_detections: 300
            }
            score_converter: SOFTMAX
          }
          second_stage_localization_loss_weight: 2.0
          second_stage_classification_loss_weight: 1.0
        }
      }
      
      train_config: {
        batch_size: 1
        optimizer {
          momentum_optimizer: {
            learning_rate: {
              manual_step_learning_rate {
                initial_learning_rate: 0.0003
                schedule {
                  step: 900000
                  learning_rate: .00003
                }
                schedule {
                  step: 1200000
                  learning_rate: .000003
                }
              }
            }
            momentum_optimizer_value: 0.9
          }
          use_moving_average: false
        }
        gradient_clipping_by_norm: 10.0
        fine_tune_checkpoint: "PATH_TO_BE_CONFIGURED/model.ckpt"
        from_detection_checkpoint: true
        # Note: The below line limits the training process to 200K steps, which we
        # empirically found to be sufficient enough to train the pets dataset. This
        # effectively bypasses the learning rate schedule (the learning rate will
        # never decay). Remove the below line to train indefinitely.
        num_steps: 200000
        data_augmentation_options {
          random_horizontal_flip {
          }
        }
        max_number_of_boxes: 50
      }
      
      train_input_reader: {
        tf_record_input_reader {
          input_path: "PATH_TO_BE_CONFIGURED/pet_train.record"
        }
        label_map_path: "PATH_TO_BE_CONFIGURED/pet_label_map.pbtxt"
      }
      
      eval_config: {
        num_examples: 2000
        # Note: The below line limits the evaluation process to 10 evaluations.
        # Remove the below line to evaluate indefinitely.
        max_evals: 10
      }
      
      eval_input_reader: {
        tf_record_input_reader {
          input_path: "PATH_TO_BE_CONFIGURED/pet_val.record"
        }
        label_map_path: "PATH_TO_BE_CONFIGURED/pet_label_map.pbtxt"
        shuffle: false
        num_readers: 1
      }

      第7至10行示意这num_classes是faster_rcnnmessage 的参数之一，而后者又是message的参数model。一样，optimizer是父train_config音讯的子音讯，而message的batch_size另一个参数train_config。我们能够经由过程签出响应的protos文件来考证这一点。

    • syntax = "proto2";
      
      package object_detection.protos;
      
      import "object_detection/protos/anchor_generator.proto";
      import "object_detection/protos/box_predictor.proto";
      import "object_detection/protos/hyperparams.proto";
      import "object_detection/protos/image_resizer.proto";
      import "object_detection/protos/losses.proto";
      import "object_detection/protos/post_processing.proto";
      
      // Configuration for Faster R-CNN models.
      // See meta_architectures/faster_rcnn_meta_arch.py and models/model_builder.py
      //
      // Naming conventions:
      // Faster R-CNN models have two stages: a first stage region proposal network
      // (or RPN) and a second stage box classifier.  We thus use the prefixes
      // `first_stage_` and `second_stage_` to indicate the stage to which each
      // parameter pertains when relevant.
      message FasterRcnn {
      
        // Whether to construct only the Region Proposal Network (RPN).
        optional int32 number_of_stages = 1 [default=2];
      
        // Number of classes to predict.
        optional int32 num_classes = 3;
        
        // Image resizer for preprocessing the input image.
        optional ImageResizer image_resizer = 4;

      从第20行和第26行能够显著看出，这num_classes是optional音讯的参数之一faster_rcnn。我愿望到目前为止的议论有助于明白设置文件的构造。如今，是时刻准确更新模子的参数之一了。

    • 步骤1：确定要更新的参数

      假定我们须要更新fast_rcnn_resnet50_pets.config文件的image_resizer第10行中提到的参数。

      步骤2：在存储库中搜刮给定参数

      目的是找到proto参数文件。为此，我们须要在存储库中搜刮。

    • 

       

       我们须要搜刮以下代码：

    • parameter_name path:research/object_detection/protos
      #in our case parameter_name="image_resizer" thus,
      image_resizer path:research/object_detection/protos

      在此path:research/object_detection/protos限定搜刮域。在此处能够找到有关怎样在GitHub上搜刮的更多信息。搜刮的输出image_resizer path:research/object_detection/protos以下所示：

    • 

    •  

      从输出中很显著，要更新image_resizer参数，我们须要剖析image_resizer.proto文件。

      步骤3：剖析proto档案

       

      syntax = "proto2";
      
      package object_detection.protos;
      
      // Configuration proto for image resizing operations.
      // See builders/image_resizer_builder.py for details.
      message ImageResizer {
        oneof image_resizer_oneof {
          KeepAspectRatioResizer keep_aspect_ratio_resizer = 1;
          FixedShapeResizer fixed_shape_resizer = 2;
        }
      }
      
      // Enumeration type for image resizing methods provided in TensorFlow.
      enum ResizeType {
        BILINEAR = 0; // Corresponds to tf.image.ResizeMethod.BILINEAR
        NEAREST_NEIGHBOR = 1; // Corresponds to tf.image.ResizeMethod.NEAREST_NEIGHBOR
        BICUBIC = 2; // Corresponds to tf.image.ResizeMethod.BICUBIC
        AREA = 3; // Corresponds to tf.image.ResizeMethod.AREA
      }
      
      // Configuration proto for image resizer that keeps aspect ratio.
      message KeepAspectRatioResizer {
        // Desired size of the smaller image dimension in pixels.
        optional int32 min_dimension = 1 [default = 600];
      
        // Desired size of the larger image dimension in pixels.
        optional int32 max_dimension = 2 [default = 1024];
      
        // Desired method when resizing image.
        optional ResizeType resize_method = 3 [default = BILINEAR];
      
        // Whether to pad the image with zeros so the output spatial size is
        // [max_dimension, max_dimension]. Note that the zeros are padded to the
        // bottom and the right of the resized image.
        optional bool pad_to_max_dimension = 4 [default = false];
      
        // Whether to also resize the image channels from 3 to 1 (RGB to grayscale).
        optional bool convert_to_grayscale = 5 [default = false];
      
        // Per-channel pad value. This is only used when pad_to_max_dimension is True.
        // If unspecified, a default pad value of 0 is applied to all channels.
        repeated float per_channel_pad_value = 6;
      }
      
      // Configuration proto for image resizer that resizes to a fixed shape.
      message FixedShapeResizer {
        // Desired height of image in pixels.
        optional int32 height = 1 [default = 300];
      
        // Desired width of image in pixels.
        optional int32 width = 2 [default = 300];
      
        // Desired method when resizing image.
        optional ResizeType resize_method = 3 [default = BILINEAR];
      
        // Whether to also resize the image channels from 3 to 1 (RGB to grayscale).
        optional bool convert_to_grayscale = 4 [default = false];
      }

      从第8-10行能够看出，我们能够运用keep_aspect_ratio_resizer或调解图象的大小fixed_shape_resizer。在剖析行23-44，我们能够观察到的音讯keep_aspect_ratio_resizer有参数：min_dimension，max_dimension，resize_method，pad_to_max_dimension，convert_to_grayscale，和per_channel_pad_value。另外，fixed_shape_resizer有参数：height，width，resize_method，和convert_to_grayscale。proto文件中提到了一切参数的数据范例。因而，要变动image_resizer范例，我们能够在设置文件中变动以下几行。

    • #before
      image_resizer {
      keep_aspect_ratio_resizer {
      min_dimension: 600 
      max_dimension: 1024
          }
      }
      #after
      image_resizer {
      fixed_shape_resizer {
      height: 600
      width: 500
      resize_method: AREA
        }
      }

      上面的代码将运用AREA调解大小要领将图象调解为500 * 600。TensorFlow中可用的种种调解大小的要领能够在这里找到。

    • 其他例子

      我们能够运用上一节中议论的步骤更新/增加任何参数。我将在此处演示一些常常运用的示例，然则上面议论的步骤能够有助于更新/增加模子的任何参数。

      变动分量初始化器

      • 决议变动fast_rcnn_resnet50_pets.config文件的initializer第35行的参数。
      • initializer path:research/object_detection/protos在存储库中搜刮。依据搜刮效果，很显著我们须要剖析hyperparams.proto文件。
      • 
      • • hyperparams.proto文件中的第68–74行说清楚明了initializer设置。

        • message Initializer {
            oneof initializer_oneof {
              TruncatedNormalInitializer truncated_normal_initializer = 1;
              VarianceScalingInitializer variance_scaling_initializer = 2;
              RandomNormalInitializer random_normal_initializer = 3;
            }
          }

          我们能够运用random_normal_intializer替代truncated_normal_initializer，由于我们须要剖析hyperparams.proto文件中的第99–102行。

        • message RandomNormalInitializer {
          optional float mean = 1 [default = 0.0];
          optional float stddev = 2 [default = 1.0];
          }
        • 显著random_normal_intializer有两个参数mean和stddev。我们能够将设置文件中的以下几行变动为use random_normal_intializer。

        • #before
          initializer {
              truncated_normal_initializer {
                  stddev: 0.01
                 }
          }
          #after
          initializer {
              random_normal_intializer{
                 mean: 1 
                 stddev: 0.5
                 }
          }

          变动体重优化器

          • 决议变动faster_rcnn_resnet50_pets.config文件的第87行momentum_optimizer的父音讯的参数。optimizer
          • optimizer path:research/object_detection/protos在存储库中搜刮。依据搜刮效果，很显著我们须要剖析optimizer.proto文件。
          • 
          • • optimizer.proto文件中的9-14行，诠释optimizer设置。

             

            message Optimizer {
              oneof optimizer {
                RMSPropOptimizer rms_prop_optimizer = 1;
                MomentumOptimizer momentum_optimizer = 2;
                AdamOptimizer adam_optimizer = 3;
              }

            显著，替代momentum_optimizer我们能够运用adam_optimizer已被证实是优越的优化顺序。为此，我们须要在f aster_rcnn_resnet50_pets.config文件中举行以下变动。

         

        #before
        optimizer {  
          momentum_optimizer: {
              learning_rate: {
                   manual_step_learning_rate {
                  initial_learning_rate: 0.0003
                  schedule {
                    step: 900000
                    learning_rate: .00003
                  }
                  schedule {
                    step: 1200000
                    learning_rate: .000003
                  }
                }
              }
              momentum_optimizer_value: 0.9
            }
        #after
        optimizer {
          adam_optimizer: {
              learning_rate: {
               manual_step_learning_rate {
                  initial_learning_rate: 0.0003
                  schedule {
                    step: 900000
                    learning_rate: .00003
                  }
                  schedule {
                    step: 1200000
                    learning_rate: .000003
                  }
                }
              }
            }

        评价预练习模子

        Eval守候300秒，以搜检练习模子是不是已更新！假如您的GPU不错，那末您能够同时举行练习和评价！一般，资本将被耗尽。为了战胜这个题目，我们能够先练习模子，将其保存在目次中，然后再评价模子。为了稍后举行评价，我们须要在设置文件中举行以下变动：

      • #Before
        eval_config: {
          num_examples: 2000
          # Note: The below line limits the evaluation process to 10 evaluations.
          # Remove the below line to evaluate indefinitely.
          max_evals: 10
        }
        #after
        eval_config: {
        num_examples: 10
        num_visualizations: 10
        eval_interval_secs: 0
        }

        num_visualizations应当即是要评价的数目！可视化的数目越多，评价所需的时候就越多。假如您的GPU具有充足的才能同时举行练习和评价，则能够保存eval_interval_secs: 300。此参数决议运转评价的频次。我根据上面议论的3个步骤得出了这个结论。

        简而言之，协定缓冲区的学问协助我们明白了模子参数是以音讯情势通报的，而且能够更新我们能够援用的.proto文件的参数。议论了3个简朴的步骤来找到.proto用于更新参数的准确文件。

        请在解释的设置文件中说起您要更新/增加的任何参数。

      • 关注【OpenCV与AI深度进修】取得更多资讯

        扫描下面二维码即可关注

      •