时间戳

首页 > 解决方案 > 无法从检查点加载模型

python - 无法从检查点加载模型

问题描述

当我使用默认命令开始训练时，使用模型动物园中的对象检测 API 和 mask-rcnn：

python model_main_tf2.py --pipeline_config_path=pipeline.config --model_dir=model/v1/

我收到这个错误

Two checkpoint references resolved to different objects (<tensorflow.python.keras.layers.convolutional.Conv2D object at 0x7f5e94075c10> and <tensorflow.python.keras.layers.convolutional.Conv2D object at 0x7f5e9403b710>).
Traceback (most recent call last):
  File "model_main_tf2.py", line 115, in <module>
    tf.compat.v1.app.run()
  File "/home/.local/lib/python3.7/site-packages/tensorflow/python/platform/app.py", line 40, in run
    _run(main=main, argv=argv, flags_parser=_parse_flags_tolerate_undef)
  File "/opt/anaconda/envs/tensorflow2.3/lib/python3.7/site-packages/absl/app.py", line 300, in run
    _run_main(main, args)
  File "/opt/anaconda/envs/tensorflow2.3/lib/python3.7/site-packages/absl/app.py", line 251, in _run_main
    sys.exit(main(argv))
  File "/home/TF/workspace/Antenna/model_main_tf2.py", line 112, in main
    record_summaries=FLAGS.record_summaries)
  File "/home/TF/models/research/object_detection/model_lib_v2.py", line 603, in train_loop
    train_input, unpad_groundtruth_tensors)
  File "/home/TF/models/research/object_detection/model_lib_v2.py", line 401, in load_fine_tune_checkpoint
    checkpoint_path).expect_partial().assert_existing_objects_matched()
  File "/home/.local/lib/python3.7/site-packages/tensorflow/python/training/tracking/util.py", line 1721, in restore
    status = self._saver.restore(save_path=save_path)
  File "/home/.local/lib/python3.7/site-packages/tensorflow/python/training/tracking/util.py", line 1320, in restore
    checkpoint=checkpoint, proto_id=0).restore(self._graph_view.root)
  File "/home/.local/lib/python3.7/site-packages/tensorflow/python/training/tracking/base.py", line 209, in restore
    restore_ops = trackable._restore_from_checkpoint_position(self)  # pylint: disable=protected-access
  File "/home/.local/lib/python3.7/site-packages/tensorflow/python/training/tracking/base.py", line 914, in _restore_from_checkpoint_position
    tensor_saveables, python_saveables))
  File "/home/.local/lib/python3.7/site-packages/tensorflow/python/training/tracking/util.py", line 297, in restore_saveables
    validated_saveables).restore(self.save_path_tensor, self.options)
  File "/home/.local/lib/python3.7/site-packages/tensorflow/python/training/saving/functional_saver.py", line 340, in restore
    restore_ops = restore_fn()
  File "/home/.local/lib/python3.7/site-packages/tensorflow/python/training/saving/functional_saver.py", line 316, in restore_fn
    restore_ops.update(saver.restore(file_prefix, options))
  File "/home/.local/lib/python3.7/site-packages/tensorflow/python/training/saving/functional_saver.py", line 111, in restore
    restored_tensors, restored_shapes=None)
  File "/home/.local/lib/python3.7/site-packages/tensorflow/python/distribute/values.py", line 890, in restore
    for v in self._mirrored_variable.values))
  File "/home/.local/lib/python3.7/site-packages/tensorflow/python/distribute/values.py", line 890, in <genexpr>
    for v in self._mirrored_variable.values))
  File "/home/.local/lib/python3.7/site-packages/tensorflow/python/distribute/values_util.py", line 195, in assign_on_device
    return variable.assign(tensor)
  File "/home/.local/lib/python3.7/site-packages/tensorflow/python/ops/resource_variable_ops.py", line 858, in assign
    self._shape.assert_is_compatible_with(value_tensor.shape)
  File "/home/.local/lib/python3.7/site-packages/tensorflow/python/framework/tensor_shape.py", line 1134, in assert_is_compatible_with
    raise ValueError("Shapes %s and %s are incompatible" % (self, other))
ValueError: Shapes (1, 1, 1088, 256) and (1, 1, 1024, 512) are incompatible

我使用来自 tf-api models/research/object-detection/configs 的默认配置

我已经尝试通过使用模型动物园或fine_tune_checkpoint_type 的配置文件来修复它：“检测”

标签： pythontensorflowtensorflow2.0object-detection-api

解决方案

文件中的输入形状pipeline.config与模型检查点中的形状不匹配

推荐阅读