Sudo ln -s /usr/local/cuda/targets/x86_64-linux/lib/libcusparse.so.10 /usr/local/cuda/lib64/libcusparse.so.10.0Īnd add the following to my ~/.bashrc - export PATH=/usr/local/cuda/bin:$PATHĮxport LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATHĮxport PATH=/usr/local/cuda-10.1/bin:$PATHĮxport LD_LIBRARY_PATH=/usr/local/cuda-10. Sudo ln -s /usr/local/cuda/targets/x86_64-linux/lib/libcudart.so /usr/local/cuda/lib64/libcudart.so.10.0 This script locates the NVIDIA CUDA C tools. Sudo ln -s /usr/local/cuda/targets/x86_64-linux/lib/libcufft.so.10 /usr/local/cuda/lib64/libcufft.so.10.0 You can check your cuda version using nvcc -version cuDNN version using cat /usr/include/cudnn.h grep CUDNNMAJOR -A 2 tensorflow-gpu version using pip freeze grep tensorflow-gpu UPDATE: Since tensorflow 2.0, has been released, I will share the compatible cuda and cuDNN versions for it as well (for Ubuntu 18.04). New in version 3.17: To find and use the CUDA toolkit libraries manually, use the FindCUDAToolkit module. Sudo ln -s /usr/local/cuda/targets/x86_64-linux/lib/libcurand.so.10 /usr/local/cuda/lib64/libcurand.so.10.0 To download the CUDA Toolkit, see CUDA Toolkit Archive (NVIDIA). Sudo ln -s /usr/local/cuda/targets/x86_64-linux/lib/libcusolver.so.10 /usr/local/cuda/lib64/libcusolver.so.10.0 Install prerequisite products for GPU Coder. sudo ln -s /opt/cuda/targets/x86_64-linux/lib/libcublas.so /opt/cuda/targets/x86_64-linux/lib/libcublas.so.10.0 :īut I had to create symlinks for it to work as tensorflow originally works with CUDA 10. You can use following configurations (This worked for me - as of 9/10). I had installed CUDA 10.1 and CUDNN 7.6 by mistake. Updated as of Dec 5 2020: For the updated information please refer Link for Linux and Link for Windows. Please refer to for a up-to-date compatibility chart (for official TF wheels). The corresponding cudnn can be downloaded here. Since the given specifications below in some cases might be too broad, here is one specific configuration that works: to Video Codec SDK 8.0.14 afir audio filter scalecuda CUDA based video scale filter. The following images and the link provide an overview of the officially supported/tested combinations of CUDA and TensorFlow on Linux, macOS and Windows: Minor configurations: NVIDIA NVDEC-accelerated H.264, HEVC, MJPEG, MPEG-1/2/4, VC1. Sources: Press materials received from the company and additional information gleaned from the company’s website.Check the CUDA version: cat /usr/local/cuda/version.txtĪnd cuDNN version: grep CUDNN_MAJOR -A 2 /usr/local/cuda/include/cudnn.hĪnd install a combination as given below in the images or here. In addition, the new CUDA Toolkit 3.2 release includes H.264 encode/decode, new Tesla Compute Cluster (TCC) integration, cluster management features, and support for the new 6GB NVIDIA Tesla and Quadro GPU products. A host of additional improvements to GPU debugging and performance analysis tools.New CUSPARSE library of sparse matrix routines.Up to 300% performance improvement in CUDA BLAS (CUBLAS) library routines.The CUDA Toolkit includes all the tools, libraries and documentation developers need to build CUDA C/C++ applications, and is the foundation for many other GPU computing language solutions.Īccording to the company, new features and significant performance enhancements in version 3.2 include: NVIDIA has announced the availability of the CUDA Toolkit 3.2 production release, which provides performance increases, new math libraries and advanced cluster management features for developers creating GPU-accelerated applications.
0 Comments
Leave a Reply. |