我刚用pip更新了TF，并尝试为单个gpu系统运行cifar演示。我在运行单gpu版本的演示代码时遇到以下错误。演示代码的multi-gpu版本运行良好。此前，单gpu演示在0.5.0以下运行良好。
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)]