“tensorflow.python.framework.errors_impl.FailedPreconditionError: =; Is a directory”运行 model_main.py 出错

Question

我运行命令：

python3 research/object_detection/legacy/train.py \
    --logtostderr \
    --train_dir = training/ \
    --pipeline_config_path = training/ssd_mobilenet_v1_pets.config

但是我收到这个错误：

WARNING: The TensorFlow contrib module will not be included in TensorFlow 2.0.
For more information, please see:
  * https://github.com/tensorflow/community/blob/master/rfcs/20180907-contrib-sunset.md
  * https://github.com/tensorflow/addons
If you depend on functionality not listed there, please file an issue.
WARNING:tensorflow:From /home/apoorva/.local/lib/python3.5/site-packages/tensorflow/python/platform/app.py:125: main (from __main__) is deprecated and will be removed in a future version.
Instructions for updating:
Use object_detection/model_main.py.
Traceback (most recent call last):
  File "research/object_detection/legacy/train.py", line 184, in <module>
    tf.app.run()
  File "/home/apoorva/.local/lib/python3.5/site-packages/tensorflow/python/platform/app.py", line 125, in run
    _sys.exit(main(argv))
  File "/home/apoorva/.local/lib/python3.5/site-packages/tensorflow/python/util/deprecation.py", line 324, in new_func
    return func(*args, **kwargs)
  File "research/object_detection/legacy/train.py", line 93, in main
    FLAGS.pipeline_config_path)
  File "/home/apoorva/Tensorflowapi/models-master/research/slim/object_detection/utils/config_util.py", line 97, in get_configs_from_pipeline_file
    proto_str = f.read()
  File "/home/apoorva/.local/lib/python3.5/site-packages/tensorflow/python/lib/io/file_io.py", line 132, in read
    pywrap_tensorflow.ReadFromStream(self._read_buf, length, status))
  File "/home/apoorva/.local/lib/python3.5/site-packages/tensorflow/python/framework/errors_impl.py", line 528, in __exit__
    c_api.TF_GetCode(self.status.status))
tensorflow.python.framework.errors_impl.FailedPreconditionError: =; Is a directory

我在

train.py

中运行

models/research/object_detection/legacy

，得到同样的错误所以尝试运行

model_main.py

这是

train.py

的更高版本，但仍然无法解决问题。

python3 research/object_detection/legacy/train.py \
    --logtostderr \
    --train_dir = training/ \
    --pipeline_config_path = training/ssd_mobilenet_v1_pets.config

我希望训练开始并显示每一步的损失和准确性，但命令没有运行。

Answer 1

您正在以错误的格式调用命令。

python3 research/object_detection/legacy/train.py \
--logtostderr --train_dir=training/ \
--pipeline_config_path=training/ssd_mobilenet_v1_pets.config

--train_dir

和

=

和

training

之间不能有空格。否则

=

被解释为参数的值。另外推荐使用

model_main.py

，所以正确的调用方式是：

python3 object_detection/model_main.py --alsologtostderr \
--model_dir=training/\
--pipeline_config_path=training/ssd_mobilenet_v1_pets.config

Answer 2

我会为那些可能遇到上述错误的人写更多的细节。 上面的错误是因为上面回答者说的命令写错了

--train_dir=训练/

以上需要更正。应根据“=”删除空格。 如果去掉空格后仍然出现上述错误，需要再次检查是否将文件名或文件夹路径作为参数。