0

我正在尝试使用 Tensorflow 对象检测 API（Mask RCNN）训练实例分割模型，并按照此处的说明进行操作。

我正在使用预训练mask_rcnn_resnet50_atrous_coco来初始化权重，并为所述模型调整了这个示例配置文件。我已经根据create_coco_tf_record.py创建了带有掩码的 tfrecord 文件，用于训练和评估集。我能够成功运行训练脚本，但问题是，除了 GPU 内存之外，它还占用了45GB我的RAM. 除此之外，一切都运行良好，我能够完成最多 10k 步的训练，之后它决定需要更多 RAM 并占用60GB我的系统崩溃。当我在训练后运行评估脚本时也会发生同样的事情。以下是我的系统规格：

  • Ubuntu 16.04
  • Tensorflow 1.5.0（根据文档通过 pip 安装）
  • Python 2.7.12
  • CUDA 9 和 CuDnn 7
  • GTX 1080 (8 GB)
  • 32 GB RAM 和 32 GB 交换

当我在 GPU 上运行模型时，我不确定为什么 tensorflow 需要这么多 RAM。我只有1 foreground class和周围500最多50 objects/masks每个图像的训练样本。以下是我的文件System Monitor中的一些屏幕截图。如果需要，我还将上传我的脚本以创建 tfrecord 文件。 nvidia-smipipeline_config在此处输入图像描述 在此处输入图像描述

这是我的管道配置：

# Mask R-CNN with Resnet-50 (v1), Atrous version
# Configured 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 {
  faster_rcnn {
    num_classes: 1
    image_resizer {
      keep_aspect_ratio_resizer {
        min_dimension: 300
        max_dimension: 400
      }
    }
    number_of_stages: 3
    feature_extractor {
      type: 'faster_rcnn_resnet50'
      first_stage_features_stride: 8
    }
    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: 8
        width_stride: 8
      }
    }
    first_stage_atrous_rate: 2
    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: true
        dropout_keep_probability: 0.5
        predict_instance_masks: true
        mask_height: 33
        mask_width: 33
        mask_prediction_conv_depth: 0
        mask_prediction_num_conv_layers: 4
        fc_hyperparams {
          op: FC
          regularizer {
            l2_regularizer {
              weight: 0.0
            }
          }
          initializer {
            variance_scaling_initializer {
              factor: 1.0
              uniform: true
              mode: FAN_AVG
            }
          }
        }
        conv_hyperparams {
          op: CONV
          regularizer {
            l2_regularizer {
              weight: 0.0
            }
          }
          initializer {
            truncated_normal_initializer {
              stddev: 0.01
            }
          }
        }
      }
    }
    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
    second_stage_mask_prediction_loss_weight: 4.0
    second_stage_batch_size: 4
  }
}

train_config: {
  batch_size: 1
  optimizer {
    momentum_optimizer: {
      learning_rate: {
        manual_step_learning_rate {
          initial_learning_rate: 0.0003
          schedule {
            step: 0
            learning_rate: .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: "/media/ahmed/1A6E52446E5218B9/Projects/TF/MaskRCNN/pretrained_models/mask_rcnn_resnet50_atrous_coco_2018_01_28/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: 50000
  data_augmentation_options {
    random_horizontal_flip {
    }
  }
}

train_input_reader: {
  tf_record_input_reader {
    input_path: "/media/ahmed/1A6E52446E5218B9/Projects/TF/MaskRCNN/train_mask.record"
  }
  label_map_path: "/media/ahmed/1A6E52446E5218B9/Projects/TF/label_map.pbtxt"
  load_instance_masks: true
  mask_type: PNG_MASKS
}

eval_config: {
  num_examples: 200
  num_visualizations : 200
  # 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: "/media/ahmed/1A6E52446E5218B9/Projects/TF/MaskRCNN/val_mask.record"
  }
  label_map_path: "/media/ahmed/1A6E52446E5218B9/Projects/TF/label_map.pbtxt"
  load_instance_masks: true
  mask_type: PNG_MASKS
  shuffle: false
  num_readers: 1
}

更新 1： 我尝试在我的frozen_inference_graph. 只要我不调用掩码输出节点，它就可以运行 2 GB 的 GPU 内存。我得到正确的盒子、班级和分数。当我尝试获取 in 的输出时detection_masks:0，sess.run()我的 8GB GPU 内存不足。我尝试在 CPU 模式下运行脚本，发现我得到了正确的掩码，我的 RAM 内存使用量从未超过 8GB（从 2.5GB 开始）。同样的模型在运行train.py和eval.py脚本时占用了我的整个 GPU 以及 45GB 的 RAM。

4

0 回答 0