Faster R-CNN模型评估重复报错：忽略已添加图像ID问题求助

使用faster_rcnn_resnet101_v1_1024x1024_coco17_tpu-8预训练模型，训练过程无异常，但执行评估命令时，加载cuDNN版本8400后，持续抛出WARNING:tensorflow:Ignoring ground truth/detection with image id 1016176252 since it was previously added警告直至中断。已尝试添加num_examples和max_evals参数调整无效，另一数据集评估正常。

执行命令 #

python model_main_tf2.py --pipeline_config_path=./training_outlook_action_ctx/training_1/pipeline.config --model_dir=./training_outlook_action_ctx/training_1 --checkpoint_dir=./training_outlook_action_ctx/training_1

报错信息 #

WARNING:tensorflow:Ignoring ground truth with image id 1016176252 since it was previously added
W0810 10:17:12.131517 140545620840832 coco_evaluation.py:113] Ignoring ground truth with image id 1016176252 since it was previously added
WARNING:tensorflow:Ignoring detection with image id 1016176252 since it was previously added
W0810 10:17:12.131881 140545620840832 coco_evaluation.py:196] Ignoring detection with image id 1016176252 since it was previously added
...

pipeline.config配置 #

# Faster R-CNN with Resnet-50 (v1)
# Trained on COCO, initialized from Imagenet classification checkpoint

# This config is TPU compatible.

model {
  faster_rcnn {
    num_classes: 7
    image_resizer {
      fixed_shape_resizer {
        width: 1024
        height: 1024
      }
    }
    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, 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
            }
          }
        }
        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: 2
  sync_replicas: true
  startup_delay_steps: 0
  replicas_to_aggregate: 8
  num_steps: 200000
  optimizer {
    momentum_optimizer: {
      learning_rate: {
        cosine_decay_learning_rate {
          learning_rate_base: .04
          total_steps: 100000
          warmup_learning_rate: .013333
          warmup_steps: 2000
        }
      }
      momentum_optimizer_value: 0.9
    }
    use_moving_average: false
  }
  fine_tune_checkpoint_version: V2
  fine_tune_checkpoint: "/pretrained_models/faster_rcnn_resnet101_v1_1024x1024_coco17_tpu-8/checkpoint/ckpt-0"
  fine_tune_checkpoint_type: "detection"
  data_augmentation_options {
    random_horizontal_flip {
    }
  }

  data_augmentation_options {
    random_adjust_hue {
    }
  }

  data_augmentation_options {
    random_adjust_contrast {
    }
  }

  data_augmentation_options {
    random_adjust_saturation {
    }
  }

  data_augmentation_options {
     random_square_crop_by_scale {
      scale_min: 0.6
      scale_max: 1.3
    }
  }
  max_number_of_boxes: 100
  unpad_groundtruth_tensors: false
  use_bfloat16: true  # works only on TPUs
}
train_input_reader: {
  label_map_path: "./training_outlook_action_ctx/data/label_map.pbtxt"
  tf_record_input_reader {
    input_path: "./training_outlook_action_ctx/data/train.records"
  }
}

eval_config: {
  metrics_set: "coco_detection_metrics"
  use_moving_averages: false
  batch_size: 2
}

eval_input_reader: {
  label_map_path: "./training_outlook_action_ctx/data/label_map.pbtxt"
  shuffle: false
  tf_record_input_reader {
    input_path: "./training_outlook_action_ctx/data/train.records"
  }
}

环境信息 #

• OS: Debian GNU/Linux 11 (bullseye)
• Python: 3.9.12
• Tensorflow: 2.9.1

解决方法 #

1. 检查并修复数据集重复数据：警告核心原因是评估用的TFRecord文件中存在重复的image id，同一个ID被多次写入。需要检查生成TFRecord的脚本，确保每个图像只被处理一次，重新生成无重复数据的TFRecord。
2. 更换评估数据集：当前配置中eval_input_reader使用的是训练集train.records，建议替换为独立的测试集test.records，符合评估逻辑的同时避免重复ID问题。
3. 过滤重复记录：如果必须使用训练集评估，可编写脚本遍历TFRecord，过滤掉重复image id的记录后重新生成文件。

内容的提问来源于stack exchange，提问作者edi graovac