Question

我正在尝试使用TensorFlow对象检测api，以便将COCO14的MobileNetSSD（深度0.75）（深度学习）重新训练为1类（人）。我正在寻找帮助，以指示哪些参数对于分类最重要。

使用不同的优化程序（RMSProp，Adam），我得到相似的结果，并且从320k / 500k的步进达到了Adam的平稳状态。此外，我认为亚当会自动调整学习速度，但从total_loss图中可以清楚地看出，我的手动降低有助于学习过程。

我最近与亚当的结果类似：

哪个给出了mAP

 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.273
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.501
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.261
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.037
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.349
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.652
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.145
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.330
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.391
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.138
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.522
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.753

配置：

 optimizer {
    adam_optimizer: {
      learning_rate: {
        manual_step_learning_rate {
          initial_learning_rate: 0.001
          schedule {
            step: 100000
            learning_rate: 0.0005
          }
          schedule {
            step: 200000
            learning_rate: 0.0002
          }
          schedule {
            step: 300000
            learning_rate: 0.0001
          }
        }
      }     
    }
    use_moving_average: false
  }

此结果与我使用RmsProp进行的实验（200k步）非常相似：

 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.259
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.483
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.245
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.031
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.326
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.645
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.142
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.317
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.379
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.127
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.508
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.743

配置：

  optimizer {
    rms_prop_optimizer {
      learning_rate {
        exponential_decay_learning_rate {
          initial_learning_rate: 0.004000000189989805
          decay_steps: 20000
          decay_factor: 0.949999988079071
        }
      }
      momentum_optimizer_value: 0.8999999761581421
      decay: 0.8999999761581421
      epsilon: 1.0
    }
    use_moving_average: false
  }

对于余弦优化器，我的结果完全错误（总损耗在增加）。

似乎我应该尝试调整一些网络参数，但是哪些是最重要的？

下面是我的整个配置。数据集具有64115个训练和2693个验证示例（没有人群注释）。

model {
  ssd {
    num_classes: 1
    box_coder {
      faster_rcnn_box_coder {
        y_scale: 10.0
        x_scale: 10.0
        height_scale: 5.0
        width_scale: 5.0
      }
    }
    matcher {
      argmax_matcher {
        matched_threshold: 0.5
        unmatched_threshold: 0.5
        ignore_thresholds: false
        negatives_lower_than_unmatched: true
        force_match_for_each_row: true
      }
    }
    similarity_calculator {
      iou_similarity {
      }
    }
    anchor_generator {
      ssd_anchor_generator {
        num_layers: 6
        min_scale: 0.2
        max_scale: 0.95
        aspect_ratios: 1.0
        aspect_ratios: 2.0
        aspect_ratios: 0.5
        aspect_ratios: 3.0
        aspect_ratios: 0.3333
      }
    }
    image_resizer {
      fixed_shape_resizer {
        height: 300
        width: 300
      }
    }
    box_predictor {
      convolutional_box_predictor {
        min_depth: 0
        max_depth: 0
        num_layers_before_predictor: 0
        use_dropout: false
        dropout_keep_probability: 0.8
        kernel_size: 1
        box_code_size: 4
        apply_sigmoid_to_scores: false
        conv_hyperparams {
          activation: RELU_6,
          regularizer {
            l2_regularizer {
              weight: 0.00004
            }
          }
          initializer {
            truncated_normal_initializer {
              stddev: 0.03
              mean: 0.0
            }
          }
          batch_norm {
            train: true,
            scale: true,
            center: true,
            decay: 0.9997,
            epsilon: 0.001,
          }
        }
      }
    }
   feature_extractor {
      type: "ssd_mobilenet_v1"
      depth_multiplier: 0.75
      min_depth: 16
      conv_hyperparams {
        regularizer {
          l2_regularizer {
            weight: 3.99999989895e-05
          }
        }
        initializer {
          truncated_normal_initializer {
            mean: 0.0
            stddev: 0.0299999993294
          }
        }
        activation: RELU_6
        batch_norm {
          decay: 0.97000002861
          center: true
          scale: true
          epsilon: 0.0010000000475
          train: true
        }
      }
      override_base_feature_extractor_hyperparams: true
    }
    loss {
      classification_loss {
        weighted_sigmoid {
        }
      }
      localization_loss {
        weighted_smooth_l1 {
        }
      }
      hard_example_miner {
        num_hard_examples: 3000
        iou_threshold: 0.99
        loss_type: CLASSIFICATION
        max_negatives_per_positive: 3
        min_negatives_per_image: 0
      }
      classification_weight: 1.0
      localization_weight: 1.0
    }
    normalize_loss_by_num_matches: true
    post_processing {
      batch_non_max_suppression {
        score_threshold: 1e-8
        iou_threshold: 0.6
        max_detections_per_class: 100
        max_total_detections: 100
      }
      score_converter: SIGMOID
    }
  }
}
train_config {
  batch_size: 24
  data_augmentation_options {
    random_horizontal_flip {
    }
  }
  data_augmentation_options {
    ssd_random_crop {
    }
  }
  optimizer {
    adam_optimizer: {
      learning_rate: {
        manual_step_learning_rate {
          initial_learning_rate: 0.001
          schedule {
            step: 100000
            learning_rate: 0.0005
          }
          schedule {
            step: 200000
            learning_rate: 0.0002
          }
          schedule {
            step: 300000
            learning_rate: 0.0001
          }
        }
      }     
    }
    use_moving_average: false
  }
  fine_tune_checkpoint: "/home/tobiasz/Projects/tf-retrain-ssd/training_assets/pretrained_model/model.ckpt"
  from_detection_checkpoint: true
  load_all_detection_checkpoint_vars: false
  num_steps:500000
}
train_input_reader {
  label_map_path: "classes.pbtxt"
  tf_record_input_reader {
    input_path: "/coco_train_300.record"
  }
}
eval_config {
  num_examples: 2693
  metrics_set: "coco_detection_metrics"
  use_moving_averages: false
  num_visualizations: 20
}
eval_input_reader {
  label_map_path: "classes.pbtxt"
  shuffle: false
  num_readers: 1
  tf_record_input_reader {
    input_path: "coco_val_300.record"
  }
}

顺便说一句，我开始阅读MobileNet和SSD论文，但是您的意见将对我有很大帮助。

哪些参数对培训影响最大（MobileNet SSD）

0 个答案: