2021-03-21 12:41:45.991059: 我tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库cudart64_110.dll 2021-03-21 12:42:22.975683：我 tensorflow/compiler/jit/xla_cpu_device.cc:41] 未创建 XLA 设备，未设置 tf_xla_enable_xla_devices 2021-03-21 12:42:22.977610：我 tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库 nvcuda.dll 2021-03-21 12:42:23.968500：我 tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] 发现设备 0 具有以下属性： pciBusID：0000:01:00.0 名称：GeForce GTX 1050 计算能力：6.1 coreClock：1.493GHz coreCount：5 deviceMemorySize：4.00GiB deviceMemoryBandwidth：104.43GiB/s 2021-03-21 12:42:23.969013: 我tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库cudart64_110.dll 2021-03-21 12:42:23.979037：我 tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库 cublas64_11.dll 2021-03-21 12:42:23.979073：我 tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库 cublasLt64_11.dll 2021-03-21 12:42:23.984739：我 tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库 cufft64_10.dll 2021-03-21 12:42:23.989507: 我tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库curand64_10.dll 2021-03-21 12:42:24.000278: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库 cusolver64_10.dll 2021-03-21 12:42:24.004780：我 tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库 cusparse64_11.dll 2021-03-21 12:42:24.006742：我 tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库 cudnn64_8.dll 2021-03-21 12:42:24.006865: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1862] 添加可见 GPU 设备：0 2021-03-21 12:42:24.008350: I tensorflow/core/platform/cpu_feature_guard.cc:142] 这个 TensorFlow 二进制文件使用 oneAPI 深度神经网络库 (oneDNN) 进行了优化，以在性能关键操作中使用以下 CPU 指令： AVX2 要在其他操作中启用它们，请使用适当的编译器标志重建 TensorFlow。 2021-03-21 12:42:24.008710：我 tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] 发现设备 0 具有以下属性： pciBusID：0000:01:00.0 名称：GeForce GTX 1050 计算能力：6.1 coreClock：1.493GHz coreCount：5 deviceMemorySize：4.00GiB deviceMemoryBandwidth：104.43GiB/s 2021-03-21 12:42:24.008737: 我tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库cudart64_110.dll 2021-03-21 12:42:24.008750：我 tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库 cublas64_11.dll 2021-03-21 12:42:24.008774：我 tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库 cublasLt64_11.dll 2021-03-21 12:42:24.008784：我 tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库 cufft64_10.dll 2021-03-21 12:42:24.008794：我 tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库 curand64_10.dll 2021-03-21 12:42:24.008804: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库 cusolver64_10.dll 2021-03-21 12:42:24.008816：我 tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库 cusparse64_11.dll 2021-03-21 12:42:24.008825: 我 tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库 cudnn64_8.dll 2021-03-21 12:42:24.008890：我 tensorflow/core/common_runtime/gpu/gpu_device.cc:1862] 添加可见的 gpu 设备：0 2021-03-21 12:42:24.645909: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1261] 具有强度 1 边缘矩阵的设备互连 StreamExecutor： 2021-03-21 12:42:24.645945：我张量流/核心/common_runtime/gpu/gpu_device.cc:1267] 0 2021-03-21 12:42:24.646033: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1280] 0: N 2021-03-21 12:42:24.646310：我 tensorflow/core/common_runtime/gpu/gpu_device.cc:1406] 创建了 TensorFlow 设备（/job:localhost/replica:0/task:0/device:GPU:0 with 2989 MB 内存）-> 物理 GPU（设备：0，名称：GeForce GTX 1050，pci 总线 ID：0000:01:00.0，计算能力：6.1） 2021-03-21 12:42:24.647083：我 tensorflow/compiler/jit/xla_gpu_device.cc:99] 未创建 XLA 设备，未设置 tf_xla_enable_xla_devices 2021-03-21 12:42:40.041994: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:116] 没有启用任何 MLIR 优化通道（注册 2） 2021-03-21 12:42:59.619450：我 tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库 cublas64_11.dll 2021-03-21 12:42:59.945615：我 tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库 cublasLt64_11.dll 2021-03-21 12:42:59.977196：我 tensorflow/stream_executor/platform/default/dso_loader.cc:49] 成功打开动态库 cudnn64_8.dll 2021-03-21 12:43:29.115454: E tensorflow/stream_executor/dnn.cc:616] CUDNN_STATUS_EXECUTION_FAILED 在 tensorflow/stream_executor/cuda/cuda_dnn.cc(1972): 'cudnnRNNBackwardData(cudnn.handle(), rnn_desc.handle(), model_dims.max_seq_length, output_desc.handles(), output_data.opaque(.), output_desc , output_backprop_data.opaque(), output_h_desc.handle(), output_h_backprop_data.opaque(), output_c_desc.handle(), output_c_backprop_data.opaque(), rnn_desc.params_handle(), params.opaque(), input_handle_data(), input_h .opaque(), input_c_desc.handle(), input_c_data.opaque(), input_desc.handles(), input_backprop_data->opaque(), input_h_desc.handle(), input_h_backprop_data->opaque(), input_c_desc.handle_prop_back_prop ->opaque()、workspace.opaque()、workspace.size()、reserve_space_data->opaque()、reserve_space_data->size())' 2021-03-21 12:43:29.115503: W tensorflow/core/framework/op_kernel.cc:1763] OP_REQUIRES 在 cudnn_rnn_ops.cc:1926 失败：内部：无法调用 ThenRnnBackward 模型_输入模式，[modernn_modernn] : 2, 0, 0 , [num_layers, input_size, num_units, dir_count, max_seq_length, batch_size, cell_num_units]: [1, 256, 20, 1, 4096, 1, 20]