Using Tensorflow backend
2018-02-16 14:22:41.660690: I tensorflow/core/platform/cpu_feature_guard.cc:137] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 AVX AVX2 FMA 2018-02-16 14:22:41.825828: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1105] Found device 0 with properties: name: GeForce GTX 980 major: 5 minor: 2 memoryClockRate(GHz): 1.2155 pciBusID: 0000:02:00.0 totalMemory: 3.94GiB freeMemory: 3.75GiB 2018-02-16 14:22:41.825860: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1195] Creating TensorFlow device (/device:GPU:0) → (device: 0, name: GeForce GTX 980, pci bus id: 0000:02:00.0, compute capability: 5.2) 2018-02-16 14:22:43.182548: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1195] Creating TensorFlow device (/device:GPU:0) → (device: 0, name: GeForce GTX 980, pci bus id: 0000:02:00.0, compute capability: 5.2) 2018-02-16 14:22:43.773744: E tensorflow/stream_executor/cuda/cuda_dnn.cc:385] could not create cudnn handle: CUDNN_STATUS_NOT_INITIALIZED 2018-02-16 14:22:43.773833: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:369] driver version file contents: “”“NVRM version: NVIDIA UNIX x86_64 Kernel Module 384.111 Tue Dec 19 23:51:45 PST 2017 GCC version: gcc version 5.4.0 20160609 (Ubuntu 5.4.0-6ubuntu1~16.04.5) “”” 2018-02-16 14:22:43.773879: E tensorflow/stream_executor/cuda/cuda_dnn.cc:393] possibly insufficient driver version: 384.111.0 2018-02-16 14:22:43.773898: E tensorflow/stream_executor/cuda/cuda_dnn.cc:352] could not destroy cudnn handle: CUDNN_STATUS_BAD_PARAM 2018-02-16 14:22:43.773914: F tensorflow/core/kernels/conv_ops.cc:717] Check failed: stream->parent()->GetConvolveAlgorithms( conv_parameters.ShouldIncludeWinogradNonfusedAlgo(), &algorithms) Aborted (core dumped)
well, upgrading to a higher version of nvidia driver does help. But nvidia-387 does have some issues with linux kernel 4.13.