天天看點
傳回

could not create cudnn handle: CUDNN_STATUS_INTERNAL_ERROR could not destroy cudnn handle: CU問題解決方案

問題

2018-04-27 17:39:18.270790: E T:\src\github\tensorflow\tensorflow\stream_executor\cuda\cuda_dnn.cc:403] could not create cudnn handle: CUDNN_STATUS_INTERNAL_ERROR
2018-04-27 17:39:18.270961: E T:\src\github\tensorflow\tensorflow\stream_executor\cuda\cuda_dnn.cc:370] could not destroy cudnn handle: CUDNN_STATUS_BAD_PARAM
2018-04-27 17:39:18.271076: F T:\src\github\tensorflow\tensorflow\core\kernels\conv_ops.cc:712] Check failed: stream->parent()->GetConvolveAlgorithms( conv_parameters.ShouldIncludeWinogradNonfusedAlgo<T>(), &algorithms)

解決：

gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=)
    with tf.Session(config=tf.ConfigProto(gpu_options=gpu_options)) as sess:
深度學習 cudnn_status_internal_error
上一篇: caffe訓練出錯：Check failed: status == CUDNN_STATUS_SUCCESS (4 vs. 0) CUDNN_STATUS_INTERNAL_ERROR
下一篇: Check failed: status == CUDNN_STATUS_SUCCESS (1 vs. 0) CUDNN_STATUS_NOT_INITIALIZED

繼續閱讀