如何为自定义数据微调niftynet预训练模型

Question

我想使用niftynet预训练的分割模型来分割自定义数据。我下载了预训练权重，并修改了model_dir路径以下载一个。

但是，当我运行

python3 net_segment.py train -c /home/Container_data/config/promise12_demo_train_config.ini

我得到了下面的错误。

Caused by op 'save/Assign_17', defined at:
    File "net_segment.py", line 8, in <module>
      sys.exit(main())
    File "/home/NiftyNet/niftynet/__init__.py", line 142, in main
      app_driver.run(app_driver.app)
    File "/home/NiftyNet/niftynet/engine/application_driver.py", line 197, in run
      SESS_STARTED.send(application, iter_msg=None)
    File "/usr/local/lib/python3.5/dist-packages/blinker/base.py", line 267, in send
      for receiver in self.receivers_for(sender)]
    File "/usr/local/lib/python3.5/dist-packages/blinker/base.py", line 267, in <listcomp>
      for receiver in self.receivers_for(sender)]
    File "/home/NiftyNet/niftynet/engine/handler_model.py", line 109, in restore_model
      var_list=to_restore, save_relative_paths=True)
    File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 1102, in __init__
      self.build()
    File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 1114, in build
      self._build(self._filename, build_save=True, build_restore=True)
    File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 1151, in _build
      build_save=build_save, build_restore=build_restore)
    File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 795, in _build_internal
      restore_sequentially, reshape)
    File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 428, in _AddRestoreOps
      assign_ops.append(saveable.restore(saveable_tensors, shapes))
    File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 119, in restore
      self.op.get_shape().is_fully_defined())
    File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/ops/state_ops.py", line 221, in assign
      validate_shape=validate_shape)
    File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/ops/gen_state_ops.py", line 61, in assign
      use_locking=use_locking, name=name)
    File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/op_def_library.py", line 787, in _apply_op_helper
      op_def=op_def)
    File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/util/deprecation.py", line 488, in new_func
      return func(*args, **kwargs)
    File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/ops.py", line 3274, in create_op
      op_def=op_def)
    File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/ops.py", line 1770, in __init__
      self._traceback = tf_stack.extract_stack()
  InvalidArgumentError (see above for traceback): 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= [3,3,61,256] rhs shape= [3,3,3,61,9]
           [[node save/Assign_17 (defined at /home/NiftyNet/niftynet/engine/handler_model.py:109)  = Assign[T=DT_FLOAT, _class=["loc:@DenseVNet/conv/conv_/w"], use_locking=true, validate_shape=true, _device="/job:localhost/replica:0/task:0/device:GPU:0"](DenseVNet/conv/conv_/w, save/RestoreV2/_35)

上面的https://github.com/tensorflow/models/issues/5390链接说要添加

--initialize_last_layer = False
--last_layers_contain_logits_only = False

有人能帮我摆脱这个错误吗？

Answer 1

看起来你在最后一层遇到了问题。当您在新任务上使用预训练模型时，您可能需要更改最后一层以适应您的新需求。

为此，您应该修改配置文件，恢复除最后一层以外的所有var：vars_to_restore = ^((?!(last_layer_name)).)*$

然后设置num_classes以适应新的分段问题。

您可以在此处查看迁移学习文档：https://niftynet.readthedocs.io/en/dev/transfer_learning.html