各位大神,菜鸟求教!

系统安装、升级讨论
版面规则
我们都知道新人的确很菜,也喜欢抱怨,并且带有浓厚的Windows习惯,但既然在这里询问,我们就应该有责任帮助他们解决问题,而不是直接泼冷水、简单的否定或发表对解决问题没有任何帮助的帖子。乐于分享,以人为本,这正是Ubuntu的精神所在。
回复
sddmzhang
帖子: 5
注册时间: 2017-09-18 20:49
系统: win10+ubuntu16.04

各位大神,菜鸟求教!

#1

帖子 sddmzhang » 2017-09-30 16:04

用eclipse调试tensorflow_fcn的代码,报这个错误,请问各位大神。能不能看一下什么问题,多谢。

Traceback (most recent call last):
File "/media/zjq/work/Tensorflow/cat/tensorflow-fcn-master/test_fcn8_vgg.py", line 28, in <module>
vgg_fcn = fcn8_vgg.FCN8VGG()
File "/media/zjq/work/Tensorflow/cat/tensorflow-fcn-master/fcn8_vgg.py", line 33, in __init__
self.data_dict = np.load(vgg16_npy_path,encoding='latin1').item()
File "/usr/local/lib/python2.7/dist-packages/numpy/lib/npyio.py", line 419, in load
pickle_kwargs=pickle_kwargs)
File "/usr/local/lib/python2.7/dist-packages/numpy/lib/format.py", line 640, in read_array
array = pickle.load(fp, **pickle_kwargs)
EOFError
头像
astolia
论坛版主
帖子: 6386
注册时间: 2008-09-18 13:11

Re: 各位大神,菜鸟求教!

#2

帖子 astolia » 2017-09-30 20:22

去看看vgg16.npy文件是否可读,并且是555245036字节大小
sddmzhang
帖子: 5
注册时间: 2017-09-18 20:49
系统: win10+ubuntu16.04

Re: 各位大神,菜鸟求教!

#3

帖子 sddmzhang » 2017-10-04 21:39

astolia 写了:去看看vgg16.npy文件是否可读,并且是555245036字节大小
多谢大神,这个问题解决了,现在又有一个问题,程序运行的时候出这一堆,您能不能给看一下什么意思,有没有解决方法,多谢,顺便,祝中秋快乐!

全部结果如下:

2017-10-04 12:41:36.339363: 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.
2017-10-04 12:41:36.339404: 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.
2017-10-04 12:41:36.339414: 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.
2017-10-04 12:41:36.339422: 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.
2017-10-04 12:41:36.339429: 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.
2017-10-04 12:41:36,447 INFO Load npy file from '/root/eclipse-workspace/fcn/vgg16.npy'.
npy file loaded
Layer name: conv1_1
Layer shape: (3, 3, 3, 64)
2017-10-04 12:41:46,261 INFO Creating Summary for: conv1_1/filter
2017-10-04 12:41:46,329 INFO Creating Summary for: conv1_1/biases
Layer name: conv1_2
Layer shape: (3, 3, 64, 64)
2017-10-04 12:41:46,347 INFO Creating Summary for: conv1_2/filter
2017-10-04 12:41:46,360 INFO Creating Summary for: conv1_2/biases
Layer name: conv2_1
Layer shape: (3, 3, 64, 128)
2017-10-04 12:41:46,380 INFO Creating Summary for: conv2_1/filter
2017-10-04 12:41:46,394 INFO Creating Summary for: conv2_1/biases
Layer name: conv2_2
Layer shape: (3, 3, 128, 128)
2017-10-04 12:41:46,413 INFO Creating Summary for: conv2_2/filter
2017-10-04 12:41:46,428 INFO Creating Summary for: conv2_2/biases
Layer name: conv3_1
Layer shape: (3, 3, 128, 256)
2017-10-04 12:41:46,451 INFO Creating Summary for: conv3_1/filter
2017-10-04 12:41:46,465 INFO Creating Summary for: conv3_1/biases
Layer name: conv3_2
Layer shape: (3, 3, 256, 256)
2017-10-04 12:41:46,487 INFO Creating Summary for: conv3_2/filter
2017-10-04 12:41:46,501 INFO Creating Summary for: conv3_2/biases
Layer name: conv3_3
Layer shape: (3, 3, 256, 256)
2017-10-04 12:41:46,524 INFO Creating Summary for: conv3_3/filter
2017-10-04 12:41:46,539 INFO Creating Summary for: conv3_3/biases
Layer name: conv4_1
Layer shape: (3, 3, 256, 512)
2017-10-04 12:41:46,564 INFO Creating Summary for: conv4_1/filter
2017-10-04 12:41:46,579 INFO Creating Summary for: conv4_1/biases
Layer name: conv4_2
Layer shape: (3, 3, 512, 512)
2017-10-04 12:41:46,610 INFO Creating Summary for: conv4_2/filter
2017-10-04 12:41:46,624 INFO Creating Summary for: conv4_2/biases
Layer name: conv4_3
Layer shape: (3, 3, 512, 512)
2017-10-04 12:41:46,654 INFO Creating Summary for: conv4_3/filter
2017-10-04 12:41:46,668 INFO Creating Summary for: conv4_3/biases
Layer name: conv5_1
Layer shape: (3, 3, 512, 512)
2017-10-04 12:41:46,696 INFO Creating Summary for: conv5_1/filter
2017-10-04 12:41:46,711 INFO Creating Summary for: conv5_1/biases
Layer name: conv5_2
Layer shape: (3, 3, 512, 512)
2017-10-04 12:41:46,737 INFO Creating Summary for: conv5_2/filter
2017-10-04 12:41:46,751 INFO Creating Summary for: conv5_2/biases
Layer name: conv5_3
Layer shape: (3, 3, 512, 512)
2017-10-04 12:41:46,780 INFO Creating Summary for: conv5_3/filter
2017-10-04 12:41:46,794 INFO Creating Summary for: conv5_3/biases
Layer name: fc6
Layer shape: [7, 7, 512, 4096]
2017-10-04 12:41:47,312 INFO Creating Summary for: fc6/weights
2017-10-04 12:41:47,326 INFO Creating Summary for: fc6/biases
Layer name: fc7
Layer shape: [1, 1, 4096, 4096]
2017-10-04 12:41:47,425 INFO Creating Summary for: fc7/weights
2017-10-04 12:41:47,438 INFO Creating Summary for: fc7/biases
Layer name: fc8
Layer shape: [1, 1, 4096, 1000]
2017-10-04 12:41:47,490 INFO Creating Summary for: score_fr/weights
2017-10-04 12:41:47,504 INFO Creating Summary for: score_fr/biases
WARNING:tensorflow:From /root/eclipse-workspace/fcn/fcn8_vgg.py:138: calling argmax (from tensorflow.python.ops.math_ops) with dimension is deprecated and will be removed in a future version.
Instructions for updating:
Use the `axis` argument instead
2017-10-04 12:41:47,590 WARNING From /root/eclipse-workspace/fcn/fcn8_vgg.py:138: calling argmax (from tensorflow.python.ops.math_ops) with dimension is deprecated and will be removed in a future version.
Instructions for updating:
Use the `axis` argument instead
2017-10-04 12:41:47,610 INFO Creating Summary for: upscore2/up_filter
2017-10-04 12:41:47,633 INFO Creating Summary for: score_pool4/weights
2017-10-04 12:41:47,648 INFO Creating Summary for: score_pool4/biases
2017-10-04 12:41:47,676 INFO Creating Summary for: upscore4/up_filter
2017-10-04 12:41:47,700 INFO Creating Summary for: score_pool3/weights
2017-10-04 12:41:47,713 INFO Creating Summary for: score_pool3/biases
2017-10-04 12:41:47,737 INFO Creating Summary for: upscore32/up_filter
WARNING:tensorflow:From /root/eclipse-workspace/fcn/fcn8_vgg.py:164: calling argmax (from tensorflow.python.ops.math_ops) with dimension is deprecated and will be removed in a future version.
Instructions for updating:
Use the `axis` argument instead
2017-10-04 12:41:47,754 WARNING From /root/eclipse-workspace/fcn/fcn8_vgg.py:164: calling argmax (from tensorflow.python.ops.math_ops) with dimension is deprecated and will be removed in a future version.
Instructions for updating:
Use the `axis` argument instead
Finished building Network.
2017-10-04 12:41:47,755 WARNING Score weights are initialized random.
2017-10-04 12:41:47,755 WARNING Do not expect meaningful results.
2017-10-04 12:41:47,755 INFO Start Initializing Variabels.
Running the Network
2017-10-04 12:42:16.490909: I tensorflow/core/kernels/logging_ops.cc:79] Shape of input image: [1 368 489 3]
2017-10-04 12:42:41.712265: I tensorflow/core/kernels/logging_ops.cc:79] Shape of pool1[1 184 245 64]
2017-10-04 12:42:43.292065: I tensorflow/core/kernels/logging_ops.cc:79] Shape of pool2[1 92 123 128]
2017-10-04 12:42:44.121408: I tensorflow/core/kernels/logging_ops.cc:79] Shape of pool3[1 46 62 256]
2017-10-04 12:42:44.935444: I tensorflow/core/kernels/logging_ops.cc:79] Shape of pool4[1 23 31 512]
2017-10-04 12:42:44.935465: I tensorflow/core/kernels/logging_ops.cc:79] Shape of pool4[1 23 31 512]
2017-10-04 12:42:49.121670: W tensorflow/core/framework/op_kernel.cc:1192] Invalid argument: Conv2DCustomBackpropInput: Size of out_backprop doesn't match computed: actual = 23, computed = 12
Traceback (most recent call last):
File "/root/eclipse-workspace/fcn/test_fcn8_vgg.py", line 46, in <module>
down, up = sess.run(tensors, feed_dict=feed_dict)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 895, in run
run_metadata_ptr)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 1124, in _run
feed_dict_tensor, options, run_metadata)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 1321, in _do_run
options, run_metadata)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 1340, in _do_call
raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InvalidArgumentError: Conv2DCustomBackpropInput: Size of out_backprop doesn't match computed: actual = 23, computed = 12
[[Node: content_vgg/upscore2/conv2d_transpose = Conv2DBackpropInput[T=DT_FLOAT, data_format="NHWC", padding="SAME", strides=[1, 2, 2, 1], use_cudnn_on_gpu=true, _device="/job:localhost/replica:0/task:0/cpu:0"](content_vgg/upscore2/stack, upscore2/up_filter/read, content_vgg/score_fr/BiasAdd)]]

Caused by op u'content_vgg/upscore2/conv2d_transpose', defined at:
File "/root/eclipse-workspace/fcn/test_fcn8_vgg.py", line 32, in <module>
vgg_fcn.build(batch_images, debug=True)
File "/root/eclipse-workspace/fcn/fcn8_vgg.py", line 144, in build
ksize=4, stride=2)
File "/root/eclipse-workspace/fcn/fcn8_vgg.py", line 274, in _upscore_layer
strides=strides, padding='SAME')
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/ops/nn_ops.py", line 1048, in conv2d_transpose
name=name)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/ops/gen_nn_ops.py", line 486, in conv2d_backprop_input
data_format=data_format, name=name)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/op_def_library.py", line 767, in apply_op
op_def=op_def)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/ops.py", line 2630, 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 1204, in __init__
self._traceback = self._graph._extract_stack() # pylint: disable=protected-access

InvalidArgumentError (see above for traceback): Conv2DCustomBackpropInput: Size of out_backprop doesn't match computed: actual = 23, computed = 12
[[Node: content_vgg/upscore2/conv2d_transpose = Conv2DBackpropInput[T=DT_FLOAT, data_format="NHWC", padding="SAME", strides=[1, 2, 2, 1], use_cudnn_on_gpu=true, _device="/job:localhost/replica:0/task:0/cpu:0"](content_vgg/upscore2/stack, upscore2/up_filter/read, content_vgg/score_fr/BiasAdd)]]
回复