21:28:44.087827: W tensorflow/stream_executor/platform/default/dso_:64] Could not load dynamic library 'libcublas.so.11' dlerror: libcublas.so.11: cannot open shared object file: No such file or directory Or 21:28:44.087779: W tensorflow/stream_executor/platform/default/dso_:64] Could not load dynamic library 'libcudart.so.11.0' dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory I've tried several tensorflow and tensorflow-gpu versions, from 2.4 to 2.7, but in everyone fails with: 21:28:43.206834: W tensorflow/stream_executor/platform/default/dso_:64] Could not load dynamic library 'libcudart.so.11.0' dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory bashrc # Nvidia cuda toolkitĮxport PATH=/usr/local/cuda-11.2/bin$ The cudnn: cat /usr/local/cuda/include/cudnn_version.h | grep CUDNN_MAJOR -A 2 | 0 N/A N/A 6384 G /usr/lib/firefox/firefox 4MiB | | 0 N/A N/A 3508 G /usr/bin/gnome-shell 77MiB | | GPU GI CI PID Type Process name GPU Memory | | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. I've installed the drivers 460.106.00 and cuda 11.2 for the Nvidia GeForce RTX 3090: +-+ I know that there are several questions with similar questions but I've tried everything I found and it didn't work, so that is why I'm writting this question: It provides type-safe parameters, per feature API header, lazy function resolution, multi-context and multi-thread support, global function callbacks, meta information about the generated OpenGL binding and the OpenGL runtime, as well as multiple examples for quick-starting your projects.I've been trying to run tensorflow in my gpu for some long days but I've been not able to accomplish it. It leverages modern C++11 features like enum classes, lambdas, and variadic templates, instead of relying on macros (all OpenGL symbols are real functions and variables). Glbinding is a new, generated, cross-platform C++ binding for OpenGL which is solely based on the new xml-based OpenGL API specification (gl.xml). #include #include int main () glbinding (C++) The only fix for earlier versions is to use glewExperimental: GLEW version 2.0.0+ uses glGetStringi instead. It also doesn't fetch the function pointers. It calls glGetString(GL_EXTENSIONS), which causes GL_INVALID_ENUM on GL 3.2+ core context as soon as glewInit() is called. GLEW up to version 1.13.0 has a problem with core contexts. Initialization of GLEW 1.13.0 and earlier The latest release as of June 2021 is version 2.2.0. If you include glxext.h before glxew.h, GLEW will complain. GLEW also provides glxew.h for X windows systems. If you include wglext.h before wglew.h, GLEW will complain. GLEW also provides wglew.h which provides Windows specific GL functions (wgl functions). In fact, you shouldn't be including gl.h at all glew.h replaces it. It supports Windows, MacOS X, Linux, and FreeBSD.Īs with most other loaders, you should not include gl.h, glext.h, or any other gl related header file before glew.h, otherwise you'll get an error message that you have included gl.h before glew.h. The OpenGL Extension Wrangler library provides access to all GL entrypoints. 4 glad (Multi-Language GL/GLES/EGL/GLX/WGL Loader-Generator).1.1 Initialization of GLEW 1.13.0 and earlier.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |