标签：rom mis -- flow str instead art 链接 mpi
之前下载了cuda11.1想要使用tensorflow的gpu版本，不料调试的时候出了问题，使用tf.test.is_gpu_available()命令测试的时候出现了如下错误：
错误日志：
WARNING:tensorflow:From <ipython-input-2-17bb7203622b>:1: is_gpu_available (from tensorflow.python.framework.test_util) is deprecated and will be removed in a future version.
Instructions for updating:
Use `tf.config.list_physical_devices(‘GPU‘)` instead.
2021-02-13 23:53:43.607123: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2021-02-13 23:53:43.621244: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library nvcuda.dll
2021-02-13 23:53:43.650897: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: GeForce GTX 1050 Ti computeCapability: 6.1
coreClock: 1.62GHz coreCount: 6 deviceMemorySize: 4.00GiB deviceMemoryBandwidth: 104.43GiB/s
2021-02-13 23:53:43.658694: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cudart64_110.dll
2021-02-13 23:53:43.670802: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cublas64_11.dll
2021-02-13 23:53:43.678653: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cublasLt64_11.dll
2021-02-13 23:53:43.687182: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cufft64_10.dll
2021-02-13 23:53:43.696485: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library curand64_10.dll
2021-02-13 23:53:43.702571: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library ‘cusolver64_10.dll‘; dlerror: cusolver64_10.dll not found
2021-02-13 23:53:43.716161: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cusparse64_11.dll
2021-02-13 23:53:43.721301: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cudnn64_8.dll
2021-02-13 23:53:43.728528: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1757] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...
2021-02-13 23:53:43.820151: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1261] Device interconnect StreamExecutor with strength 1 edge matrix:
2021-02-13 23:53:43.825857: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1267] 0
2021-02-13 23:53:43.829518: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1280] 0: N
2021-02-13 23:53:43.832609: I tensorflow/compiler/jit/xla_gpu_device.cc:99] Not creating XLA devices, tf_xla_enable_xla_devices not set
Out[2]: False
使用的版本如下：
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
上述错误也表示gpu版本并不能使用，但是看到了错误
字面意思就是少cusolver64_10.dll的动态链接库，少库肯定要去文件里找
打开自己的cuda安装目录，依次打开../NVIDIA GPU Computing Toolkit\CUDA\v11.1\bin,找到问题所在了
后缀名字是11，但是我们需要的是10，尝试更改了文件为cusolver64_10.dll，运行
没想到会是这个问题
呵呵，你妹的
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
网上也有说版本不兼容的问题，降版本是一定可以解决问题的，但我的习惯是下载任何东西都尽量保持最新。。。
看后面能不能遇见啥新问题吧。。
标签：rom mis -- flow str instead art 链接 mpi
原文地址：https://www.cnblogs.com/dd110343/p/14401386.html
