Tensorflow 使用TF 0.6.0运行CIFAR-10演示时出现新错误

我刚刚用pip更新了TF，并尝试为单个gpu系统运行cifar演示。我在运行单一gpu版本的演示代码时遇到以下错误。演示代码的多gpu版本运行良好。同样，以前的单一gpu演示在0.5.0下运行良好

错误输出的一部分：

[[Node: Variable/initial_value/_80 = _HostSend[T=DT_INT32, 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_1185_Variable/initial_value", _device="/job:localhost/replica:0/task:0/gpu:0"](Variable/initial_value)]]
W tensorflow/core/common_runtime/executor.cc:1076] 0x1b33a70 Compute status: Failed precondition: Attempting to use uninitialized value conv1/weights
     [[Node: conv1/weights/_72 = _Send[T=DT_FLOAT, client_terminated=false, recv_device="/job:localhost/replica:0/task:0/gpu:0", send_device="/job:localhost/replica:0/task:0/cpu:0", send_device_incarnation=1, tensor_name="edge_341_conv1/weights", _device="/job:localhost/replica:0/task:0/cpu:0"](conv1/weights)]]
     [[Node: local4/weights/Assign/_27 = _Send[T=DT_FLOAT, client_terminated=false, recv_device="/job:localhost/replica:0/task:0/gpu:0", send_device="/job:localhost/replica:0/task:0/cpu:0", send_device_incarnation=1, tensor_name="edge_218_local4/weights/Assign", _device="/job:localhost/replica:0/task:0/cpu:0"](local4/weights/Assign/_26)]]
W tensorflow/core/common_runtime/executor.cc:1076] 0x1b33a70 Compute status: Failed precondition: Attempting to use uninitialized value conv1/weights
     [[Node: conv1/weights/_72 = _Send[T=DT_FLOAT, client_terminated=false, recv_device="/job:localhost/replica:0/task:0/gpu:0", send_device="/job:localhost/replica:0/task:0/cpu:0", send_device_incarnation=1, tensor_name="edge_341_conv1/weights", _device="/job:localhost/replica:0/task:0/cpu:0"](conv1/weights)]]
     [[Node: conv2/weights/Assign/_59 = _Send[T=DT_FLOAT, client_terminated=false, recv_device="/job:localhost/replica:0/task:0/gpu:0", send_device="/job:localhost/replica:0/task:0/cpu:0", send_device_incarnation=1, tensor_name="edge_291_conv2/weights/Assign", _device="/job:localhost/replica:0/task:0/cpu:0"](conv2/weights/Assign/_58)]]
W tensorflow/core/common_runtime/executor.cc:1076] 0x1b33a70 Compute status: Failed precondition: Attempting to use uninitialized value conv1/weights
     [[Node: conv1/weights/_72 = _Send[T=DT_FLOAT, client_terminated=false, recv_device="/job:localhost/replica:0/task:0/gpu:0", send_device="/job:localhost/replica:0/task:0/cpu:0", send_device_incarnation=1, tensor_name="edge_341_conv1/weights", _device="/job:localhost/replica:0/task:0/cpu:0"](conv1/weights)]]
     [[Node: local3/weights/Assign/_43 = _Send[T=DT_FLOAT, client_terminated=false, recv_device="/job:localhost/replica:0/task:0/gpu:0", send_device="/job:localhost/replica:0/task:0/cpu:0", send_device_incarnation=1, tensor_name="edge_260_local3/weights/Assign", _device="/job:localhost/replica:0/task:0/cpu:0"](local3/weights/Assign/_42)]]
Traceback (most recent call last):
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/models/image/cifar10/cifar10_train.py", line 138, in <module>
    tf.app.run()
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/platform/default/_app.py", line 30, in run
    sys.exit(main(sys.argv))
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/models/image/cifar10/cifar10_train.py", line 134, in main
    train()
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/models/image/cifar10/cifar10_train.py", line 94, in train
    sess.run(init)
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 368, in run
    results = self._do_run(target_list, unique_fetch_targets, feed_dict_string)
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", line 444, in _do_run
    e.code)
tensorflow.python.framework.errors.FailedPreconditionError: Attempting to use uninitialized value conv1/weights
     [[Node: conv1/weights/_72 = _Send[T=DT_FLOAT, client_terminated=false, recv_device="/job:localhost/replica:0/task:0/gpu:0", send_device="/job:localhost/replica:0/task:0/cpu:0", send_device_incarnation=1, tensor_name="edge_341_conv1/weights", _device="/job:localhost/replica:0/task:0/cpu:0"](conv1/weights)]

[[Node:Variable/initial\u value/\u 80=\u HostSend[T=DT\u INT32，client\u terminated=false，recv\u device=“/job:localhost/replica:0/task:0/cpu:0”，send\u device=“/job:localhost/replica:0/task:0/gpu:0”，send\u device\u化身=1，tensor\u name=“edge\u Variable/initial\u value”，\u device=“/job:localhost/replica:0/task:0/gpu:0]”（Variable/initial\u value）]
W tensorflow/core/common_runtime/executor.cc:1076]0x1b33a70计算状态：失败的前提条件：尝试使用未初始化的值conv1/weights
[[Node:conv1/weights/\u 72=\u Send[T=DT\u FLOAT，client\u terminated=false，recv\u device=“/job:localhost/replica:0/gpu:0”，Send\u device=“/job:localhost/replica:0/task:0/cpu:0”，Send\u device\u化身=1，tensor\u name=“edge\u 341\u conv1/weights”，“\u device=“/job:localhost/replica:0/task:0/cpu:0]”（conv1/weights）]]
[[Node:local4/weights/Assign/_27=\u Send[T=DT\u FLOAT，client\u terminated=false，recv\u device=“/job:localhost/replica:0/task:0/gpu:0”，Send\u device=“/job:localhost/replica:0/task:0/cpu:0”，Send\u device\u化身=1，tensor\u name=“edge\u 218\u local4/weights/Assign”，\u device=“/job:localhost/replica:0/task:0/cpu:0]”（local4/weights/Assign/]
W tensorflow/core/common_runtime/executor.cc:1076]0x1b33a70计算状态：失败的前提条件：尝试使用未初始化的值conv1/weights
[[Node:conv1/weights/\u 72=\u Send[T=DT\u FLOAT，client\u terminated=false，recv\u device=“/job:localhost/replica:0/gpu:0”，Send\u device=“/job:localhost/replica:0/task:0/cpu:0”，Send\u device\u化身=1，tensor\u name=“edge\u 341\u conv1/weights”，“\u device=“/job:localhost/replica:0/task:0/cpu:0]”（conv1/weights）]]
[[Node:conv2/weights/Assign/\u 59=\u Send[T=DT\u FLOAT，client\u terminated=false，recv\u device=“/job:localhost/replica:0/task:0/gpu:0”，Send\u device=“/job:localhost/replica:0/task:0/cpu:0”，Send\u device\u化身=1，tensor\u name=“edge\u 291\u conv2/weights/Assign”，“\u device=“/job:localhost/replica:0/task:0/cpu:0”]（conv2/weights/Assign）/58）]]
W tensorflow/core/common_runtime/executor.cc:1076]0x1b33a70计算状态：失败的前提条件：尝试使用未初始化的值conv1/weights
[[Node:conv1/weights/\u 72=\u Send[T=DT\u FLOAT，client\u terminated=false，recv\u device=“/job:localhost/replica:0/gpu:0”，Send\u device=“/job:localhost/replica:0/task:0/cpu:0”，Send\u device\u化身=1，tensor\u name=“edge\u 341\u conv1/weights”，“\u device=“/job:localhost/replica:0/task:0/cpu:0]”（conv1/weights）]]
[[Node:local3/weights/Assign/\u 43=\u Send[T=DT\u FLOAT，client\u terminated=false，recv\u device=“/job:localhost/replica:0/task:0/gpu:0”，Send\u device=“/job:localhost/replica:0/task:0/cpu:0”，Send\u device\u化身=1，tensor\u name=“edge\u 260\u local3/weights/Assign”“，\u device=“/job:localhost/replica:0/task:0/cpu:0]”（local3/weights/Assign/\u 42）]]
回溯（最近一次呼叫最后一次）：
文件“/usr/local/lib/python2.7/dist packages/tensorflow/models/image/cifar10/cifar10_train.py”，第138行，在
tf.app.run（）
文件“/usr/local/lib/python2.7/dist-packages/tensorflow/python/platform/default/_-app.py”，第30行，正在运行
系统出口（主（系统argv））
文件“/usr/local/lib/python2.7/dist packages/tensorflow/models/image/cifar10/cifar10_train.py”，第134行，主目录
列车（）
文件“/usr/local/lib/python2.7/dist packages/tensorflow/models/image/cifar10/cifar10_train.py”，第94行，列车中
sess.run（初始化）
文件“/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py”，第368行，正在运行
结果=self.\u do\u run（目标\u列表、唯一\u获取\u目标、提要\u dict\u字符串）
文件“/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py”，第444行，在
e、 代码）
tensorflow.python.framework.errors.FailedPremissionError:尝试使用未初始化的值conv1/weights
[[Node:conv1/weights/\u 72=\u Send[T=DT\u FLOAT，client\u terminated=false，recv\u device=“/job:localhost/replica:0/gpu:0”，Send\u device=“/job:localhost/replica:0/task:0/cpu:0”，Send\u device\u化身=1，tensor\u name=“edge\u 341\u conv1/weights”，“\u device=“/job:localhost/replica:0/task:0/cpu:0]”（conv1/weights）]

FYI，这是一个已知错误，将在下一版本中修复：