我正在测试使用tf.nn.conv2d的TensorFlow程序，但发生如下错误：
2018-09-18 01:33:54.908161: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
2018-09-18 01:33:54.987724: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:897] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2018-09-18 01:33:54.988106: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1405] Found device 0 with properties:
name: GeForce GTX 1070 major: 6 minor: 1 memoryClockRate(GHz): 1.695
pciBusID: 0000:01:00.0
totalMemory: 7.92GiB freeMemory: 7.44GiB
2018-09-18 01:33:54.988122: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1484] Adding visible gpu devices: 0
2018-09-18 01:33:55.193045: I tensorflow/core/common_runtime/gpu/gpu_device.cc:965] Device interconnect StreamExecutor with strength 1 edge matrix:
2018-09-18 01:33:55.193076: I tensorflow/core/common_runtime/gpu/gpu_device.cc:971] 0
2018-09-18 01:33:55.193082: I tensorflow/core/common_runtime/gpu/gpu_device.cc:984] 0: N
2018-09-18 01:33:55.193257: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1097] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 7173 MB memory) -> physical GPU (device: 0, name: GeForce GTX 1070, pci bus id: 0000:01:00.0, compute capability: 6.1)
Traceback (most recent call last):
File "/home/jzsb/.local/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1278, in _do_call
return fn(*args)
File "/home/jzsb/.local/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1263, in _run_fn
options, feed_dict, fetch_list, target_list, run_metadata)
File "/home/jzsb/.local/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1350, in _call_tf_sessionrun
run_metadata)
tensorflow.python.framework.errors_impl.UnimplementedError: Conv2D for GPU is not currently supported without cudnn
[[Node: Conv2D = Conv2D[T=DT_FLOAT, data_format="NCHW", dilations=[1, 1, 1, 1], padding="SAME", strides=[1, 1, 1, 1], use_cudnn_on_gpu=false, _device="/job:localhost/replica:0/task:0/device:GPU:0"](gradients/Conv2D_grad/Conv2DBackpropFilter-0-TransposeNHWCToNCHW-LayoutOptimizer, Variable/read)]]
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "facerec.py", line 77, in <module>
l, _ = sess.run([loss, train], feed_dict={x:imag, y:ans})
File "/home/jzsb/.local/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 877, in run
run_metadata_ptr)
File "/home/jzsb/.local/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1100, in _run
feed_dict_tensor, options, run_metadata)
File "/home/jzsb/.local/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1272, in _do_run
run_metadata)
File "/home/jzsb/.local/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1291, in _do_call
raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.UnimplementedError: Conv2D for GPU is not currently supported without cudnn
[[Node: Conv2D = Conv2D[T=DT_FLOAT, data_format="NCHW", dilations=[1, 1, 1, 1], padding="SAME", strides=[1, 1, 1, 1], use_cudnn_on_gpu=false, _device="/job:localhost/replica:0/task:0/device:GPU:0"](gradients/Conv2D_grad/Conv2DBackpropFilter-0-TransposeNHWCToNCHW-LayoutOptimizer, Variable/read)]]
Caused by op 'Conv2D', defined at:
File "facerec.py", line 31, in <module>
l1 = tf.nn.relu(tf.nn.conv2d(x, kcore1, [1, 1, 1, 1], 'SAME', False)+bias1)
File "/home/jzsb/.local/lib/python3.6/site-packages/tensorflow/python/ops/gen_nn_ops.py", line 956, in conv2d
data_format=data_format, dilations=dilations, name=name)
File "/home/jzsb/.local/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py", line 787, in _apply_op_helper
op_def=op_def)
File "/home/jzsb/.local/lib/python3.6/site-packages/tensorflow/python/util/deprecation.py", line 454, in new_func
return func(*args, **kwargs)
File "/home/jzsb/.local/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 3155, in create_op
op_def=op_def)
File "/home/jzsb/.local/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 1717, in __init__
self._traceback = tf_stack.extract_stack()
UnimplementedError (see above for traceback): Conv2D for GPU is not currently supported without cudnn
[[Node: Conv2D = Conv2D[T=DT_FLOAT, data_format="NCHW", dilations=[1, 1, 1, 1], padding="SAME", strides=[1, 1, 1, 1], use_cudnn_on_gpu=false, _device="/job:localhost/replica:0/task:0/device:GPU:0"](gradients/Conv2D_grad/Conv2DBackpropFilter-0-TransposeNHWCToNCHW-LayoutOptimizer, Variable/read)]]
我肯定会确认之前已经安装了cuDNN。
jzsb@jzsb-tf:~/PycharmProjects/untitled$ ls -las /usr/local/cuda/include/*dnn*
100 -r--r--r-- 1 root root 100962 8月 15 01:08 /usr/local/cuda/include/cudnn.h
jzsb@jzsb-tf:~/PycharmProjects/untitled$ ls -las /usr/local/cuda/lib64/*dnn*
0 lrwxrwxrwx 1 root root 13 8月 15 01:09 /usr/local/cuda/lib64/libcudnn.so -> libcudnn.so.7
0 lrwxrwxrwx 1 root root 17 8月 15 01:09 /usr/local/cuda/lib64/libcudnn.so.7 -> libcudnn.so.7.2.1
281260 -rwxr-xr-x 1 root root 288007984 8月 15 01:09 /usr/local/cuda/lib64/libcudnn.so.7.2.1
278240 -rw-r--r-- 1 root root 284914492 8月 15 01:09 /usr/local/cuda/lib64/libcudnn_static.a
我的tensorflow版本是1.10，请告诉我如何解决此问题，谢谢。