如何在ssd_mobilenet_v1张量流中增加num_classes

debugcn 发表于 Dev

林哲圭

我正在使用ssd_mobilenet_v1_coco.config和

在计划培训后添加13件事后，我将num_classes的值更改为20

python model_main.py --alsologtostderr --model_dir=training/ --pipeline_config_path=training/ssd_mobilenet_v1_coco.config

我一直尝试学习该命令，但出现错误。增加num_classes我应该怎么做？我应该从头开始抓起num_classes = 100吗？我需要帮助。

model {
  ssd {
    num_classes: 20
    box_coder {
      faster_rcnn_box_coder {
        y_scale: 10.0
        x_scale: 10.0
        height_scale: 5.0
        width_scale: 5.0
      }


  File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/training/saver.py", line 1326, in restore
    err, "a mismatch between the current graph and the graph")
tensorflow.python.framework.errors_impl.InvalidArgumentError: Restoring from checkpoint failed. This is most likely due to a mismatch between the current graph and the graph from the checkpoint. Please ensure that you have not altered the graph expected based on the checkpoint. Original error:

Assign requires shapes of both tensors to match. lhs shape= [126] rhs shape= [84]
         [[node save/Assign_56 (defined at /usr/local/lib/python3.6/dist-packages/tensorflow_core/python/framework/ops.py:1748) ]]

吉尔赫姆·贾奇琴

我最近有一个类似的问题。为了解决我的问题，我必须执行以下操作：

在pipeline.config的train_config部分中，将fine_tune_checkpoint指向先前的模型检查点。例如：`fine_tune_checkpoint：“ ./model/model.ckpt”
在model_main.py命令调用中，使model_dir指向与先前检查点不同的文件夹：

python research/object_detection/model_main.py \
  --model_dir=./model/finetune0 \
  --pipeline_config_path=./model/pipeline.config \
  --alsologtostderr

我的文件结构：

+ models
-+ model
--+ checkpoint
--+ model.ckpt.index
--+ model.ckpt.meta
--+ model.ckpt.data-00000-of-00001
--+ pipeline.config
--- finetune0 (will be autogenerated)

-- data (tfrecord dataset)
-- annotations (labels)
...