Tensorflow 2对象检测API低mAP

问题描述

我正在尝试使用Tensorflow 2.0对象检测训练更快的r-cnn模型,但是我得到的mAP极低,为0.01。

我在Tensorboard中查看了训练图像,但训练图像看起来无法正确加载,或者我在配置文件中做错了什么。 。我正在使用Hardhat示例数据集关注RoboFlow教程。这是我的colab笔记本(https://colab.research.google.com/drive/1cjHpLYq8NAEce36mJGGg0Lec31wSdtF9?usp=sharing)。

顶部图像显示了已在Tensorboard中加载的训练数据集中使用的图像,而下面的图像是原始图像。

Training Image loaded in tensorboard

Original Image in Roboflow Hard Hat Sample

对此我是完全陌生的,我不确定哪里出了问题。下面是我正在使用的配置文件

model {
  faster_rcnn {
    num_classes: 3
    image_resizer {
      keep_aspect_ratio_resizer {
        min_dimension: 640
        max_dimension: 640
        pad_to_max_dimension: true
      }
    }
    feature_extractor {
      type: 'faster_rcnn_resnet101_keras'
      batch_norm_trainable: true
    }
    first_stage_anchor_generator {
      grid_anchor_generator {
        scales: [0.25,0.5,1.0,2.0]
        aspect_ratios: [0.5,2.0]
        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
            }
          }
        }
        share_Box_across_classes: true
      }
    }
    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
    use_static_shapes: true
    use_matmul_crop_and_resize: true
    clip_anchors_to_image: true
    use_static_balanced_label_sampler: true
    use_matmul_gather_in_matcher: true
  }
}

train_config: {
  batch_size: 1
  sync_replicas: true
  startup_delay_steps: 0
  replicas_to_aggregate: 8
  num_steps: 2000
  optimizer {
    momentum_optimizer: {
      learning_rate: {
        cosine_decay_learning_rate {
          learning_rate_base: .04
          total_steps: 25000
          warmup_learning_rate: .013333
          warmup_steps: 2000
        }
      }
      momentum_optimizer_value: 0.9
    }
    use_moving_average: false
  }
  fine_tune_checkpoint_version: V2
  fine_tune_checkpoint: "/content/models/research/deploy/faster_rcnn_resnet101_v1_640x640_coco17_tpu-8/checkpoint/ckpt-0"
  fine_tune_checkpoint_type: "detection"
  data_augmentation_options {
    random_horizontal_flip {
    }
  }

  max_number_of_Boxes: 100
  unpad_groundtruth_tensors: false
  use_bfloat16: true  # works only on TPUs
}

train_input_reader: {
  label_map_path: "/content/train/Workers_label_map.pbtxt"
  tf_record_input_reader {
    input_path: "/content/train/Workers.tfrecord"
  }
}

eval_config: {
  metrics_set: "coco_detection_metrics"
  use_moving_averages: false
  batch_size: 1;
}

eval_input_reader: {
  label_map_path: "/content/train/Workers_label_map.pbtxt"
  shuffle: false
  num_epochs: 1
  tf_record_input_reader {
    input_path: "/content/valid/Workers.tfrecord"
  }
}

提前谢谢!

解决方法

看看您的训练输出,看来您应该尝试以下其中一项:

  1. 您使用的数据集只有100张图像。尝试通过augmentation(或在这种情况下,使用using the full dataset来增加7000张图像)的大小。
  2. 每个纪元花费的时间不到一秒钟,因此您的总训练时间少于5分钟。您可以尝试将num_steps从2,000提高到20,000或100,000。根据先前的经验,这些TF2模型趋向于收敛一段时间。
  3. 尝试使用不同的模型(由于YOLOv5,我已经看到YOLOv4built in augmentations在较小的数据集上收敛得更快)。