tensorflow对象检测从现有检查点微调模型

Question

我正在尝试按照这些说明从现有检查点训练模型 .

我已使用faster_rcnn_resnet101_voc07.config配置配置了对象检测培训管道.

在检查点部分,我已经设置了预定义模型的检查点文件所在的目录fast_rcnn_resnet101_coco.tar.gz

根据这个问题,fine_tune_checkpoint可以是包含三个文件的目录的路径:(.data-00000-of-00001,.index,.meta).

所以我设置了目录" / home/docs/car_dataset/models/model/train " 的路径

gradient_clipping_by_norm: 10.0
  fine_tune_checkpoint: "/home/docs/car_dataset/models/model/train"
  from_detection_checkpoint: true
  num_steps: 800000
  data_augmentation_options {
    random_horizontal_flip {
    }
  }

但是当我执行训练脚本时:

python object_detection/train.py     --logtostderr\
--pipeline_config_path=/home/docs/car_dataset/models/model/faster_rcnn_resnet101_voc07.config\
--train_dir=/home/docs/car_dataset/models/model/train\
--num_gpus=2

我收到了错误:

tensorflow.python.framework.errors_impl.DataLossError: Unable to open table file /home/docs/car_dataset/models/model/train: Failed precondition: /home/docs/car_dataset/models/model/train: perhaps your file is in a different file format and you need to use a different restore operator?

我也尝试过设置目录中每个文件的路径

fine_tune_checkpoint: "/home/docs/car_dataset/models/model/train/model.ckpt.meta"

但我得到错误:

tensorflow.python.framework.errors_impl.DataLossError: Unable to open table file /home/docs/car_dataset/models/model/train/model.ckpt.meta: Data loss: not an sstable (bad magic number): perhaps your file is in a different file format and you need to use a different restore operator?

在具有三个文件的管道配置中定义预训练模型的正确方法是什么:(.data-00000-of-00001,.index,.meta).

Tensorflow版本: 1.2.1

Answer 1

你要做的是指定没有".meta",".index"和".data-00000-of-00001"扩展名的整个路径.在您的情况下,这看起来是:"/ home/docs/car_dataset/model/model/train/model.ckpt"(您会注意到它比目录更具体).