我想在 Linux 上安装 tensorflow戴尔 Inspiron 580与NVidia GTX660显卡，所以我需要安装 CUDA 和 cuDNN，但它们不仅不起作用，而且我甚至无法正确获取驱动程序。我尝试了许多系统配置： - Ubuntu 16.04.5 服务器：{nvidia-396、nvidia-390、nvidia-340} - Ubuntu 16.04.1 服务器：{nvidia-390、nvidia-340} - Ubuntu 14.04 服务器：{nvidia-340、nvidia-304} - Ubuntu 12.04 服务器甚至无法安装驱动程序
Ubuntu 16.04 搭配 nvidia-304 运行良好，但效果不佳 - 最大可用分辨率为 1366x768，并且出现奇怪的黑框，尽管我使用的是全高清显示器。此外，CUDA 9.2 无法运行，而对于适合此驱动程序的 CUDA 5.0 版本，我找不到在最新版 Ubuntu 上安装它的方法。
我尝试从官方存储库安装 nvidia 驱动程序，也尝试从 ppa:graphics-drivers/ppa 以及从官方 NVidia 页面的 .run 文件安装。这台机器运行的是标准的 American Megatrends BIOS，因此不存在 EFI 和 SecureBoot 相关的问题。
我满足上述所有要求这里。
我尝试严格按照这些教程进行操作（始终从全新系统安装开始）：查理的博客，普吉特系统
在 nouveau 驱动程序上，无论我选择哪种桌面环境，它都能完美显示：全高清，没有任何错误。此外，tty 看起来更漂亮。使用 nvidia 驱动程序时，所有内容都处于低分辨率，DE 甚至不会出现。我只看到 上闪烁的光标/dev/tty7。
我该怎么做才能让它适用于 tensorflow？我可以更改发行版、软件配置等。我只坚持使用当前硬件配置。此外，它必须是 Linux。如果可以工作，它甚至可以是 CUDA 5.0。
下面是我使用的诊断命令：
$ sudo lspci | grep -i nvidia
01:00.0 VGA compatible controller: NVIDIA Corporation GK106 [GeForce GTX 660] (rev a1)
01:00.1 Audio device: NVIDIA Corporation GK106 HDMI Audio Controller (rev a1)
$ uname -m && cat /etc/*release
x86_64
DISTRIB_ID=Ubuntu
DISTRIB_RELEASE=17.10
DISTRIB_CODENAME=artful
DISTRIB_DESCRIPTION="Ubuntu 17.10"
NAME="Ubuntu"
VERSION="17.10 (Artful Aardvark)"
ID=ubuntu
ID_LIKE=debian
PRETTY_NAME="Ubuntu 17.10"
VERSION_ID="17.10"
HOME_URL="https://www.ubuntu.com/"
SUPPORT_URL="https://help.ubuntu.com/"
BUG_REPORT_URL="https://bugs.launchpad.net/ubuntu/"
PRIVACY_POLICY_URL="https://www.ubuntu.com/legal/terms-and-policies/privacy-policy"
VERSION_CODENAME=artful
UBUNTU_CODENAME=artful
$ gcc --version
gcc (Ubuntu 7.2.0-8ubuntu3.2) 7.2.0
Copyright (C) 2017 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
$ uname -r
4.13.0-46-generic
$ sudo lshw -c video
*-display
description: VGA compatible controller
product: GK106 [GeForce GTX 660]
vendor: NVIDIA Corporation
physical id: 0
bus info: pci@0000:01:00.0
version: a1
width: 64 bits
clock: 33MHz
capabilities: pm msi pciexpress vga_controller bus_master cap_list rom
configuration: driver=nvidia latency=0
resources: irq:16 memory:fa000000-faffffff memory:d8000000-dfffffff memory:d6000000-d7ffffff ioport:ec00(size=128) memory:c0000-dffff
$ dmesg | grep NVRM
[ 4.441575] NVRM: loading NVIDIA UNIX x86_64 Kernel Module 384.130 Wed Mar 21 03:37:26 PDT 2018 (using threaded interrupts)
[ 9.338846] NVRM: failed to copy vbios to system memory.
[ 9.339139] NVRM: RmInitAdapter failed! (0x30:0xffff:661)
[ 9.339164] NVRM: rm_init_adapter failed for device bearing minor number 0
[ 27.130008] NVRM: failed to copy vbios to system memory.
[ 27.130401] NVRM: RmInitAdapter failed! (0x30:0xffff:661)
[ 27.130516] NVRM: rm_init_adapter failed for device bearing minor number 0
[ 28.086782] NVRM: failed to copy vbios to system memory.
[ 28.087155] NVRM: RmInitAdapter failed! (0x30:0xffff:661)
[ 28.087195] NVRM: rm_init_adapter failed for device bearing minor number 0
[ 28.999220] NVRM: failed to copy vbios to system memory.
[ 28.999665] NVRM: RmInitAdapter failed! (0x30:0xffff:661)
[ 28.999735] NVRM: rm_init_adapter failed for device bearing minor number 0
[ 29.914304] NVRM: failed to copy vbios to system memory.
[ 29.914735] NVRM: RmInitAdapter failed! (0x30:0xffff:661)
[ 29.914798] NVRM: rm_init_adapter failed for device bearing minor number 0
[ 30.827998] NVRM: failed to copy vbios to system memory.
[ 30.828413] NVRM: RmInitAdapter failed! (0x30:0xffff:661)
[ 30.828544] NVRM: rm_init_adapter failed for device bearing minor number 0
$ nvidia-smi
No devices were found