python - Tensorflow GPU 检查失败 : stream‑>parent()‑>GetConvolveAlgorithms

Question

我正在尝试使用 Keras 和 Tensorflow 后端来训练我的 CNN 模型。

问题是，当我运行 fit_generator() 函数时，Spyder 中的 Python 内核崩溃了。为了添加一点上下文，我刚刚按照此 link 中的描述安装了 CUDA 和 tensorflow-gpu，以便我可以使用我的 GPU。在此之前，一切都很好。

这是完整的日志

启动内核时发生错误

2018󈚧󈚪 20:44:44.791399: I C:\tf_jenkins\workspace\rel‑win\M\windows‑gpu\PY\36\tensorflow\core\platform\cpu_feature_guard.cc:140] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
2018󈚧󈚪 20:44:45.084153: I C:\tf_jenkins\workspace\rel‑win\M\windows‑gpu\PY\36\tensorflow\core\common_runtime\gpu\gpu_device.cc:1212] Found device 0 with properties: 
name: GeForce GTX 1050 major: 6 minor: 1 memoryClockRate(GHz): 1.493
pciBusID: 0000:01:00.0
totalMemory: 4.00GiB freeMemory: 3.30GiB
2018󈚧󈚪 20:44:45.086132: I C:\tf_jenkins\workspace\rel‑win\M\windows‑gpu\PY\36\tensorflow\core\common_runtime\gpu\gpu_device.cc:1312] Adding visible gpu devices: 0
2018󈚧󈚪 20:44:45.906189: I C:\tf_jenkins\workspace\rel‑win\M\windows‑gpu\PY\36\tensorflow\core\common_runtime\gpu\gpu_device.cc:993] Creating TensorFlow device (/device:GPU:0 with 3033 MB memory) ‑> physical GPU (device: 0, name: GeForce GTX 1050, pci bus id: 0000:01:00.0, compute capability: 6.1)
2018󈚧󈚪 20:47:25.845646: I C:\tf_jenkins\workspace\rel‑win\M\windows‑gpu\PY\36\tensorflow\core\common_runtime\gpu\gpu_device.cc:1312] Adding visible gpu devices: 0
2018󈚧󈚪 20:47:25.846108: I C:\tf_jenkins\workspace\rel‑win\M\windows‑gpu\PY\36\tensorflow\core\common_runtime\gpu\gpu_device.cc:993] Creating TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 74 MB memory) ‑> physical GPU (device: 0, name: GeForce GTX 1050, pci bus id: 0000:01:00.0, compute capability: 6.1)
2018󈚧󈚪 20:47:26.499846: E C:\tf_jenkins\workspace\rel‑win\M\windows‑gpu\PY\36\tensorflow\stream_executor\cuda\cuda_dnn.cc:385] could not create cudnn handle: CUDNN_STATUS_NOT_INITIALIZED
2018󈚧󈚪 20:47:26.500247: E C:\tf_jenkins\workspace\rel‑win\M\windows‑gpu\PY\36\tensorflow\stream_executor\cuda\cuda_dnn.cc:389] error retrieving driver version: Unimplemented: kernel reported driver version not implemented on Windows
2018󈚧󈚪 20:47:26.500717: F C:\tf_jenkins\workspace\rel‑win\M\windows‑gpu\PY\36\tensorflow\core\kernels\conv_ops.cc:717] Check failed: stream‑>parent()‑>GetConvolveAlgorithms( conv_parameters.ShouldIncludeWinogradNonfusedAlgo(), &algorithms) 
2018󈚧󈚪 20:52:22.359428: E C:\tf_jenkins\workspace\rel‑win\M\windows‑gpu\PY\36\tensorflow\stream_executor\cuda\cuda_dnn.cc:385] could not create cudnn handle: CUDNN_STATUS_NOT_INITIALIZED
2018󈚧󈚪 20:52:22.359982: E C:\tf_jenkins\workspace\rel‑win\M\windows‑gpu\PY\36\tensorflow\stream_executor\cuda\cuda_dnn.cc:389] error retrieving driver version: Unimplemented: kernel reported driver version not implemented on Windows
2018󈚧󈚪 20:52:22.360678: F C:\tf_jenkins\workspace\rel‑win\M\windows‑gpu\PY\36\tensorflow\core\kernels\conv_ops.cc:717] Check failed: stream‑>parent()‑>GetConvolveAlgorithms( conv_parameters.ShouldIncludeWinogradNonfusedAlgo(), &algorithms) 

有没有人遇到过这个问题？

提前致谢

Answer 1

我刚刚遇到了同样的问题并找到了 solution in a Github issue .您需要在安装 CUDA 9.0 之后更新您的 GPU 驱动程序，因为看起来安装程序正在将您恢复到旧驱动程序。他们还建议重新启动，但就我而言，这不是必需的。