![]() ![]() (my installation path was: D:\Programs\圆4\Nvidia\Cuda_v_10_0\Development)ī) add environment variables: system variables / path must have:ĭ:\Programs\圆4\Nvidia\Cuda_v_10_0\Development\binĭ:\Programs\圆4\Nvidia\Cuda_v_10_0\Development\libnvvpĭ:\Programs\圆4\Nvidia\Cuda_v_10_0\Development\extras\CUPTI\lib圆4ĭ:\Programs\圆4\Nvidia\Cuda_v_10_0\Development\includeĪ) download cuDNN SDK v7.4 (needs registration, but it is simple) Select: CUDA Toolkit 10.0 and download base installer (2 GB) See, that tensorflow_gpu-1.13.1 needs: CUDA Toolkit v10.0, cuDNN SDK v7.4 * it is written for linux, but worked in my case # 3.6.4 |Anaconda custom (64-bit)| (default, Jan 16 2018, 10:22:32) ī) find right versions of CUDA Toolkit and cuDNN SDK for your tf version ![]() result - "GeForce GTX 1060 Compute Capability = 6.1"įind versions of CUDA Toolkit and cuDNN SDK, that you need.Tensorflow needs Compute Capability 3.5 or higher. my PC: GeForce GTX 1060 notebook (driver version - 419.35), windows 10, jupyter notebook.# good output must be => Ĭheck if your card can work with tensorflow (optional) ![]() Local_device_protos = device_lib.list_local_devices()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |