在 Linux 上运行 Tensorflow 代码时出错

Question

I ran my code on a Windows system and it worked fine. I am now trying to run it on an Ubuntu machine and I get the following error:

W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE3 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use FMA instructions, but these are available on your machine and could speed up CPU computations.
E tensorflow/core/common_runtime/executor.cc:594] Executor failed to create kernel. Invalid argument: NodeDef mentions attr 'dilations' not in Op<name=Conv2D; signature=input:T, filter:T -> output:T; attr=T:type,allowed=[DT_HALF, DT_FLOAT, DT_DOUBLE]; attr=strides:list(int); attr=use_cudnn_on_gpu:bool,default=true; attr=padding:string,allowed=["SAME", "VALID"]; attr=data_format:string,default="NHWC",allowed=["NHWC", "NCHW"]>; NodeDef: import/MobilenetV1/MobilenetV1/Conv2d_0/convolution = Conv2D[T=DT_FLOAT, data_format="NHWC", dilations=[1, 1, 1, 1], padding="SAME", strides=[1, 2, 2, 1], use_cudnn_on_gpu=true, _device="/job:localhost/replica:0/task:0/cpu:0"](_recv_import/input_0, import/MobilenetV1/Conv2d_0/weights)
     [[Node: import/MobilenetV1/MobilenetV1/Conv2d_0/convolution = Conv2D[T=DT_FLOAT, data_format="NHWC", dilations=[1, 1, 1, 1], padding="SAME", strides=[1, 2, 2, 1], use_cudnn_on_gpu=true, _device="/job:localhost/replica:0/task:0/cpu:0"](_recv_import/input_0, import/MobilenetV1/Conv2d_0/weights)]]
Traceback (most recent call last):
  File "/user/HS127/sc00858/Documents/tensorflow-for-poets-2/scripts/test.py", line 118, in <module>
    {input_operation.outputs[0]: t})
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 767, in run
    run_metadata_ptr)
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 965, in _run
    feed_dict_string, options, run_metadata)
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 1015, in _do_run
    target_list, options, run_metadata)
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 1035, in _do_call
    raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InvalidArgumentError: NodeDef mentions attr 'dilations' not in Op<name=Conv2D; signature=input:T, filter:T -> output:T; attr=T:type,allowed=[DT_HALF, DT_FLOAT, DT_DOUBLE]; attr=strides:list(int); attr=use_cudnn_on_gpu:bool,default=true; attr=padding:string,allowed=["SAME", "VALID"]; attr=data_format:string,default="NHWC",allowed=["NHWC", "NCHW"]>; NodeDef: import/MobilenetV1/MobilenetV1/Conv2d_0/convolution = Conv2D[T=DT_FLOAT, data_format="NHWC", dilations=[1, 1, 1, 1], padding="SAME", strides=[1, 2, 2, 1], use_cudnn_on_gpu=true, _device="/job:localhost/replica:0/task:0/cpu:0"](_recv_import/input_0, import/MobilenetV1/Conv2d_0/weights)
     [[Node: import/MobilenetV1/MobilenetV1/Conv2d_0/convolution = Conv2D[T=DT_FLOAT, data_format="NHWC", dilations=[1, 1, 1, 1], padding="SAME", strides=[1, 2, 2, 1], use_cudnn_on_gpu=true, _device="/job:localhost/replica:0/task:0/cpu:0"](_recv_import/input_0, import/MobilenetV1/Conv2d_0/weights)]]

Caused by op u'import/MobilenetV1/MobilenetV1/Conv2d_0/convolution', defined at:
  File "/user/HS127/sc00858/Documents/tensorflow-for-poets-2/scripts/test.py", line 85, in <module>
    graph = load_graph(model_file)
  File "/user/HS127/sc00858/Documents/tensorflow-for-poets-2/scripts/test.py", line 38, in load_graph
    tf.import_graph_def(graph_def)
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/importer.py", line 288, in import_graph_def
    op_def=op_def)
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/ops.py", line 2327, in create_op
    original_op=self._default_original_op, op_def=op_def)
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/ops.py", line 1226, in __init__
    self._traceback = _extract_stack()

InvalidArgumentError (see above for traceback): NodeDef mentions attr 'dilations' not in Op<name=Conv2D; signature=input:T, filter:T -> output:T; attr=T:type,allowed=[DT_HALF, DT_FLOAT, DT_DOUBLE]; attr=strides:list(int); attr=use_cudnn_on_gpu:bool,default=true; attr=padding:string,allowed=["SAME", "VALID"]; attr=data_format:string,default="NHWC",allowed=["NHWC", "NCHW"]>; NodeDef: import/MobilenetV1/MobilenetV1/Conv2d_0/convolution = Conv2D[T=DT_FLOAT, data_format="NHWC", dilations=[1, 1, 1, 1], padding="SAME", strides=[1, 2, 2, 1], use_cudnn_on_gpu=true, _device="/job:localhost/replica:0/task:0/cpu:0"](_recv_import/input_0, import/MobilenetV1/Conv2d_0/weights)
     [[Node: import/MobilenetV1/MobilenetV1/Conv2d_0/convolution = Conv2D[T=DT_FLOAT, data_format="NHWC", dilations=[1, 1, 1, 1], padding="SAME", strides=[1, 2, 2, 1], use_cudnn_on_gpu=true, _device="/job:localhost/replica:0/task:0/cpu:0"](_recv_import/input_0, import/MobilenetV1/Conv2d_0/weights)]]

I don't understand why this error is occurring and any help in debugging would be appreciated. My code is trying to run inference on a pre-trained mobilenet graph.

EDIT

Minimum of TF 1.5 is needed to infer from Mobilenet

Answer 1

I'm going to put my money on a Tensorflow version mismatch. Based on the following error text:

NodeDef mentions attr 'dilations' not in Op

Dilated convolutions are a rather newish concept, and likely to be changing from minor version to minor version, and it sounds like your code is utilizing it in a function, perhaps in the layers api. Check that you're running the same tensorflow version on both systems. If they're not the same version, look for the term dilations in your code as the likely culprit of change.