Question

我尝试更改cifar10示例以实现3d CNN。但是除了卷积和汇集之外，似乎还有其他一些我应该改变的事情。当我只更改了卷积层，池层和输入时，我得到以下错误。有什么建议我应该做什么，或者有什么例子吗？感谢。

 File "/mnt/disk1/Fu/Function/Tensorflow/Test/cifar10.py", line 152, in distorted_inputs
    batch_size=FLAGS.batch_size)
  File "/mnt/disk1/Fu/Function/Tensorflow/Test/cifar10_input.py", line 125, in distorted_inputs
    shuffle = False)
  File "/mnt/disk1/Fu/Function/Tensorflow/Test/cifar10_input.py", line 88, in _generate_image_and_label_batch
    tf.image_summary('images', images)
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/ops/logging_ops.py", line 151, in image_summary
    tag=tag, tensor=tensor, max_images=max_images, name=scope)
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/ops/gen_logging_ops.py", line 161, in _image_summary
    name=name)
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/op_def_library.py", line 749, in apply_op
    op_def=op_def)
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/ops.py", line 2380, 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 1298, in __init__
    self._traceback = _extract_stack()

InvalidArgumentError (see above for traceback): Tensor must be 4-D with last dim 1, 3, or 4, not [20,32,32,32,1]
         [[Node: ImageSummary = ImageSummary[T=DT_FLOAT, bad_color=Tensor<type: uint8 shape: [4] values: 255 0 0...>, max_images=3, _device="/job:localhost/replica:0/task:0/cpu:0"](ImageSummary/tag, batch)]]
         [[Node: local4/weight_loss/_104 = _Send[T=DT_FLOAT, client_terminated=false, recv_device="/job:localhost/replica:0/task:0/cpu:0", send_device="/job:localhost/replica:0/task:0/gpu:0", send_device_incarnation=1, tensor_name="edge_48_local4/weight_loss", _device="/job:localhost/replica:0/task:0/gpu:0"](local4/weight_loss)]]

Answer 1

您还需要更改图像摘要以支持3D图像（或者，更轻松地将其从图表中删除）。

张量流中的3d CNN的示例。 InvalidArgumentError（请参见上面的回溯）：Tensor必须是4-D，最后一个dim为1,3或4

1 个答案: