[cuDNN]執行模型訓練時,出現以下錯誤

    E tensorflow/stream_executor/cuda/cuda_dnn.cc:343] Loaded runtime CuDNN library: 7.1.4 but source was compiled with: 7.2.1.  CuDNN library major and minor version needs to match or have higher minor version in case of CuDNN 7.0 or later version. If using a binary install, upgrade your CuDNN library.  If building from sources, make sure the library loaded at runtime is compatible with the version specified during compile configuration.

    解決:一開始我選擇安裝 cuDNN v7.1.4,不過執行模型訓練時,程式異常結束且出現上述錯誤,大致原因是:所使用的 Tensorflow 是以 cuDNN 7.2.1 版本編譯的,需安裝相同或更高版本的 cuDNN Library。

    最後我改裝 cuDNN 7.3.1 就正常了。

    標籤 (Edit tags)
    • No tags
    您必須 登入 才能發佈評論。
    Powered by MindTouch Core