tensorflow - 获取卷积算法失败。这可能是因为 cuDNN 初始化失败，所以尝试查看是否有警告 [Op:Conv2D]

Question

我正在配置一个 GitHub 存储库，下面提到了一些要求。不过，我得到一个错误：对于 GeForce RTX 2070

cudatoolkit=10.1.243=h6bb024c_0
cudnn=7.6.5=cuda10.1_0
tensorboard=1.14.0=py36hf484d3e_0
tensorflow=1.14.0=gpu_py36h3fb9ad6_0
tensorflow-base=1.14.0=gpu_py36he45bfe2_0
tensorflow-estimator=1.14.0=py_0
tensorflow-gpu=1.14.0=h0d30ee6_0

回溯（最近一次通话最后）：

  File "/home/khawar/Desktop/AAAIFace/train_main.py", line 60, in <module>
    losses, loss_grad, tape = network.lossFunc(img_batch, age_batch, gender_batch, AIM_model)
  File "/home/khawar/Desktop/AAAIFace/network.py", line 139, in lossFunc
    z_enc = AIM.encoder(img)
  File "/home/khawar/Desktop/AAAIFace/modeleag.py", line 126, in __call__
    res = self.forward(x, *args, **kwargs)
  File "/home/khawar/Desktop/AAAIFace/network.py", line 23, in forward
    x = self.c1(x)
  File "/home/khawar/Desktop/AAAIFace/modeleag.py", line 126, in __call__
    res = self.forward(x, *args, **kwargs)
  File "/home/khawar/Desktop/AAAIFace/modeleag.py", line 143, in forward
    x = self.conv(x)
  File "/home/khawar/Desktop/AAAIFace/layers2.py", line 78, in __call__
    return self._deploy()
  File "/home/khawar/Desktop/AAAIFace/layers2.py", line 153, in _deploy
    out = tf.nn.conv2d(self.x, self.W, self.stride, self.pad, dilations=self.dilation_rate)
  File "/home/khawar/anaconda3/envs/AAAIFace/lib/python3.6/site-packages/tensorflow/python/ops/nn_ops.py", line 1953, in conv2d
    name=name)
  File "/home/khawar/anaconda3/envs/AAAIFace/lib/python3.6/site-packages/tensorflow/python/ops/gen_nn_ops.py", line 1039, in conv2d
    _six.raise_from(_core._status_to_exception(e.code, message), None)
  File "<string>", line 3, in raise_from
tensorflow.python.framework.errors_impl.UnknownError: Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above. [Op:Conv2D]

标签： tensorflowconvolution