首页 > 解决方案 > 这对于使用 Tensorflow 对象检测 API 训练的机器学习模型是否正常?

问题描述

张量板

我是机器学习的新手。我正在尝试使用 Tensorflow 对象检测 API 训练手部检测模型。我正在使用自我手数据集。训练中有 4400 张图像,测试中有 400 张图像。我正在使用 TensorFlow 1.15。我非常关注 sentdex 教程并进行了一些更改。我正在使用模型动物园的 ssd_mobilenet_v2_coco。

5000 步后损失在 1-2 之间。现在它近20000步,loss还在1-2之间。是正常的还是不正常的?我应该训练更多吗?

如果你想看一下,这是我的配置文件。

# SSD with Mobilenet v2 configuration for MSCOCO 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 {
  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_v2'
      min_depth: 16
      depth_multiplier: 1.0
      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,
        }
      }
    }
    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: 3
      }
      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
  optimizer {
    rms_prop_optimizer: {
      learning_rate: {
        exponential_decay_learning_rate {
          initial_learning_rate: 0.004
          decay_steps: 800720
          decay_factor: 0.95
        }
      }
      momentum_optimizer_value: 0.9
      decay: 0.9
      epsilon: 1.0
    }
  }
  fine_tune_checkpoint: "ssd_mobilenet_v2_coco_2018_03_29/model.ckpt"
  fine_tune_checkpoint_type: 'detection'
  # 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 {
    }
  }
  data_augmentation_options {
    ssd_random_crop {
    }
  }
}

train_input_reader: {
  tf_record_input_reader {
     input_path: "data/train-ego.record"
  }
 label_map_path: "data/hands_label_map.pbtxt"
}

eval_config: {
  num_examples: 8000
  # 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: "data/test-ego.record"
  }
   label_map_path: "data/hands_label_map.pbtxt"
  shuffle: false
  num_readers: 1
}

标签: pythontensorflowmachine-learningobject-detection-api

解决方案


在学习算法的第一个时期看到损失值大幅下降是很正常的。这是因为反向传播的工作方式。

您基本上在梯度方向(损失函数的导数)移动或多或少固定大小(学习率),具体取决于您优化模型的方式(随机梯度下降、RMSprop、Adam 等)。

但是,如果您的模型已经收敛,则意味着您的优化算法“卡”在了损失函数的局部最小值中。例如,它无法进一步探索,因为 larning 率太小而无法“退出”该区域。

如果您对结果感到满意,您可以决定停止学习,(在 keras 中,如果您的模型没有改进,则有提前停止的回调)

如果你不是,有很多不同的可能解决方案,涉及超参数的调整:

  • 改变优化算法和学习率
  • 改变网络架构,从而改变损失函数的格局
  • 做特征选择,特征工程等
  • 如果您过度拟合,则进行正则化
  • 如果你过拟合就退出
  • ETC

无论如何,这取决于你的问题。有数千个因素需要考虑


推荐阅读