问题描述
Tensorflow-gpu 绑定到特定版本的 Nvidia CUDA Toolkit 的原因是什么?当前版本似乎专门寻找 9.0，并且不适用于更高的版本.例如，我安装了最新的 Toolkit 9.2 并将其添加到路径中，但 Tensorflow-gpu 无法使用它并抱怨它正在寻找 9.0.
What is the reasoning that Tensorflow-gpu is bound to a specific version of Nvidia's CUDA Toolkit? The current version appears to look for 9.0 specifically and will not work with anything greater. For example I installed the latest Toolkit 9.2 and added it to path but Tensorflow-gpu will not work with it and complains that it is looking for 9.0.
我可以看到不支持主要版本更新而是次要版本?
I can see major version updates not being supported but a minor release?
推荐答案
这是个好问题.根据英伟达的网站，
CUDA 驱动程序向后兼容，这意味着针对特定 CUDA 版本编译的应用程序将继续适用于后续(更高版本)的驱动程序版本.
因此从技术上讲，支持 CUDA 驱动程序的后续迭代应该不成问题.在实践中，您会在网络 [1]，[2].更容易安装，从 conda 安装的 tensorflow-gpu 包目前与 CUDA 9.2 捆绑在一起.
So technically, it should not be a problem to support later iterations of a CUDA driver. And in practice, you will find working non-official pre-built binaries with later versions of CUDA and CuDNN on the net [1], [2]. Even easier to install, the tensorflow-gpu package installed from conda currently comes bundled with CUDA 9.2.
当被问及这个话题时，开发人员回答，
When asked on the topic, a dev answered,
为什么是 9.1 所需的驱动程序问题的答案，我们在 cuda 9.1 中需要的新功能不多，还有一些小问题.
所以这个原因看起来相当模糊——他可能是说 CUDA 9.1(和 9.2)需要的显卡驱动程序可能有点太新，不太方便，但这是一个没有受过教育的猜测.
So the reason looks rather vague -- he might mean that CUDA 9.1 (and 9.2) requires graphics card driver that are perhaps a bit too recent to be really convenient, but that is an uneducated guess.
如果 NVidia 在二进制兼容性方面是正确的，您可以尝试简单地将您的 CUDA 9.2 库重命名或链接为 CUDA 9.0 库，它应该可以工作.但我会在尝试之前保存我所有的工作......而且人们甚至重新编译 tensorflow 以支持更高版本的 CUDA 版本这一事实可能暗示了这将如何结束.
If NVidia is right about binary compatibility, you may try to simply rename or link your CUDA 9.2 library as a CUDA 9.0 library and it should work. But I would save all my work before attempting this... and the fact that people go as far as recompiling tensorflow to support later CUDA versions may be a hint on how this could end.
这篇关于支持 Nvidia CUDA Toolkit 9.2的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持！