Tensorflow：Cuda 计算能力 3.0。所需的最低 Cuda 能力为 3.5

Question

我正在从源代码(documentation) 安装 tensorflow。

Cuda 驱动版本：

nvcc: NVIDIA (R) Cuda compiler driver
Cuda compilation tools, release 7.5, V7.5.17

当我运行以下命令时：

bazel-bin/tensorflow/cc/tutorials_example_trainer --use_gpu

它给了我以下错误：

I tensorflow/stream_executor/dso_loader.cc:108] successfully opened CUDA library libcublas.so locally
I tensorflow/stream_executor/dso_loader.cc:108] successfully opened CUDA library libcudnn.so locally
I tensorflow/stream_executor/dso_loader.cc:108] successfully opened CUDA library libcufft.so locally
I tensorflow/stream_executor/dso_loader.cc:108] successfully opened CUDA library libcuda.so.1 locally
I tensorflow/stream_executor/dso_loader.cc:108] successfully opened CUDA library libcurand.so locally
I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:925] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
I tensorflow/core/common_runtime/gpu/gpu_init.cc:118] Found device 0 with properties: 
name: GeForce GT 640
major: 3 minor: 0 memoryClockRate (GHz) 0.9015
pciBusID 0000:05:00.0
Total memory: 2.00GiB
Free memory: 1.98GiB
I tensorflow/core/common_runtime/gpu/gpu_init.cc:138] DMA: 0 
I tensorflow/core/common_runtime/gpu/gpu_init.cc:148] 0:   Y 
I tensorflow/core/common_runtime/gpu/gpu_device.cc:843] Ignoring gpu device (device: 0, name: GeForce GT 640, pci bus id: 0000:05:00.0) with Cuda compute capability 3.0. The minimum required Cuda capability is 3.5.
I tensorflow/core/common_runtime/gpu/gpu_device.cc:843] Ignoring gpu device (device: 0, name: GeForce GT 640, pci bus id: 0000:05:00.0) with Cuda compute capability 3.0. The minimum required Cuda capability is 3.5.
I tensorflow/core/common_runtime/gpu/gpu_device.cc:843] Ignoring gpu device (device: 0, name: GeForce GT 640, pci bus id: 0000:05:00.0) with Cuda compute capability 3.0. The minimum required Cuda capability is 3.5.
I tensorflow/core/common_runtime/gpu/gpu_device.cc:843] Ignoring gpu device (device: 0, name: GeForce GT 640, pci bus id: 0000:05:00.0) with Cuda compute capability 3.0. The minimum required Cuda capability is 3.5.
I tensorflow/core/common_runtime/gpu/gpu_device.cc:843] Ignoring gpu device (device: 0, name: GeForce GT 640, pci bus id: 0000:05:00.0) with Cuda compute capability 3.0. The minimum required Cuda capability is 3.5.
I tensorflow/core/common_runtime/gpu/gpu_device.cc:843] Ignoring gpu device (device: 0, name: GeForce GT 640, pci bus id: 0000:05:00.0) with Cuda compute capability 3.0. The minimum required Cuda capability is 3.5.
I tensorflow/core/common_runtime/gpu/gpu_device.cc:843] Ignoring gpu device (device: 0, name: GeForce GT 640, pci bus id: 0000:05:00.0) with Cuda compute capability 3.0. The minimum required Cuda capability is 3.5.
I tensorflow/core/common_runtime/gpu/gpu_device.cc:843] Ignoring gpu device (device: 0, name: GeForce GT 640, pci bus id: 0000:05:00.0) with Cuda compute capability 3.0. The minimum required Cuda capability is 3.5.
I tensorflow/core/common_runtime/gpu/gpu_device.cc:843] Ignoring gpu device (device: 0, name: GeForce GT 640, pci bus id: 0000:05:00.0) with Cuda compute capability 3.0. The minimum required Cuda capability is 3.5.
I tensorflow/core/common_runtime/gpu/gpu_device.cc:843] Ignoring gpu device (device: 0, name: GeForce GT 640, pci bus id: 0000:05:00.0) with Cuda compute capability 3.0. The minimum required Cuda capability is 3.5.
F tensorflow/cc/tutorials/example_trainer.cc:128] Check failed: ::tensorflow::Status::OK() == (session->Run({{"x", x}}, {"y:0", "y_normalized:0"}, {}, &outputs)) (OK vs. Invalid argument: Cannot assign a device to node 'Cast': Could not satisfy explicit device specification '/gpu:0' because no devices matching that specification are registered in this process; available devices: /job:localhost/replica:0/task:0/cpu:0
     [[Node: Cast = Cast[DstT=DT_FLOAT, SrcT=DT_INT32, _device="/gpu:0"](Const)]])
F tensorflow/cc/tutorials/example_trainer.cc:128] Check failed: ::tensorflow::Status::OK() == (session->Run({{"x", x}}, {"y:0", "y_normalized:0"}, {}, &outputs)) (OK vs. Invalid argument: Cannot assign a device to node 'Cast': Could not satisfy explicit device specification '/gpu:0' because no devices matching that specification are registered in this process; available devices: /job:localhost/replica:0/task:0/cpu:0
     [[Node: Cast = Cast[DstT=DT_FLOAT, SrcT=DT_INT32, _device="/gpu:0"](Const)]])
F tensorflow/cc/tutorials/example_trainer.cc:128] Check failed: ::tensorflow::Status::OK() == (session->Run({{"x", x}}, {"y:0", "y_normalized:0"}, {}, &outputs)) (OK vs. Invalid argument: Cannot assign a device to node 'Cast': Could not satisfy explicit device specification '/gpu:0' because no devices matching that specification are registered in this process; available devices: /job:localhost/replica:0/task:0/cpu:0
     [[Node: Cast = Cast[DstT=DT_FLOAT, SrcT=DT_INT32, _device="/gpu:0"](Const)]])
F tensorflow/cc/tutorials/example_trainer.cc:128] Check failed: ::tensorflow::Status::OK() == (session->Run({{"x", x}}, {"y:0", "y_normalized:0"}, {}, &outputs)) (OK vs. Invalid argument: Cannot assign a device to node 'Cast': Could not satisfy explicit device specification '/gpu:0' because no devices matching that specification are registered in this process; available devices: /job:localhost/replica:0/task:0/cpu:0
     [[Node: Cast = Cast[DstT=DT_FLOAT, SrcT=DT_INT32, _device="/gpu:0"](Const)]])
Aborted (core dumped)

我需要不同的 gpu 来运行它吗？

Answer 1

对于 TensorFlow 2.1.0

我能够通过编译 TF2.1.0 的源代码在 Windows 上对其进行管理。由于 XLA 原因，TF 2.2.0 构建失败，即使为 bazel 禁用了所有 XLA 标志。还要小心使用更新的 Python 版本 - 我在使用 Python 3.8 的预构建 pip 包中遇到了一些奇怪的错误，所以我使用 Python 3.6 来解决这个问题。

一个警告 - 构建完成几个小时后，我开始使用该库，一个仅持续几秒钟的简单模型训练效果很好，但基本卷积网络的训练在 0 或 1 个 epochs 后失败了到 CUDA 错误。您的里程可能会有所不同。

Answer 2

感谢您提供 WHL！当我为编译它而奋斗了好几天（没有成功）时，我现在终于能够使用 TF，因为我的笔记本电脑只支持 Compute 3.0。在全新安装 Ubuntu 18.04 时，我无法按照您的说明进行编译，我想指出几点：

• 在您的“依赖项”部分，libjasper 不再独立可用，ffmpeg 不再从您列出的存储库中可用，并且 libtiff5-dev 不再可用（我认为有一个新版本）。我知道这主要是针对我也使用的 OpenCV 的东西。您还重复了几个包，例如 git 和 unzip。
• 在您的“Nvidia 驱动程序”部分，我认为存储库中没有该驱动程序。至少我拉不下来。使用您构建的 WHL 文件，我使用的是 Nvidia 网站上的 418 驱动程序，这似乎运行良好。
• 在“为 CUDA 9.0 安装 cudnn 7.1.4”部分中，您“cd /usr/lib/x86_64-linux-gnu”，但文件位于 /usr/local/cuda。它是否正确？我猜这些链接至少必须被告知指向 cuda 文件夹。
• 在“为 CUDA 9.0 安装 NCCL 2.2.12”部分中，您使用的是 2.2.12，但您的命令行均引用 2.1.15
• 在您的 Bazel 安装部分，您说要使用 Bazel Darwin 安装程序，但我认为这适用于 Mac。我认为您需要 Bazel Linux 安装程序。

再次感谢您为此所做的所有工作！

附：我能够通过按照这些说明对 Tensorflow 1.12 进行 git checkout 并通过使用 Bazel 0.15.0 使用 CUDA 9.2、CUDNN 7.1.4 和 NCCL 2、2、13 安装 keras_applications 和 keras_preprocessing 来构建它。有些人指出 CUDA 9.0 不能用 gcc6/g++6 编译。显然9.2可以。

Answer 3

在 anaconda 中，具有 cudatoolkit=9.0 的 tensorflow-gpu=1.12 与具有 3.0 计算能力的 gpu 兼容。这是创建新环境和安装 3.0 gpus 所需库的 c 命令。

conda create -n tf-gpu
conda activate tf-gpu
conda install tensorflow-gpu=1.12
conda install cudatoolkit=9.0

那么你可以通过以下方式尝试。

>python
import tensorflow as tf
tf.Session()

这是我的输出

名称：GeForce GT 650M 主要：3 次要：0 memoryClockRate(GHz)：0.95 pciBusID: 0000:01:00.0 总内存：3.94GiB 免费内存：3.26GiB 2019-12-09 13:26:11.753591：我 tensorflow/core/common_runtime/gpu/gpu_device.cc:1511] 添加可见 gpu 设备：0 2019-12-09 13:26:12.050152：我 tensorflow/core/common_runtime/gpu/gpu_device.cc:982] 设备互连 StreamExecutor 与强度 1 边缘矩阵： 2019-12-09 13:26:12.050199：我 tensorflow/core/common_runtime/gpu/gpu_device.cc:988] 0 2019-12-09 13:26:12.050222: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1001] 0: N 2019-12-09 13:26:12.050481：我 tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] 创建了 TensorFlow 设备（/job:localhost/replica:0/task:0/device:GPU:0 和 2989 MB 内存）-> 物理 GPU（设备：0，名称：GeForce GT 650M，pci 总线 ID：0000:01:00.0，计算能力：3.0）

享受吧！

Answer 4

@Taako，很抱歉这么晚才回复。我没有保存上面显示的编译的轮文件。但是，这是 tensorflow 1.9 的新版本。希望这对您有足够的帮助。请确保以下用于构建的详细信息。

张量流：1.9 CUDA 工具包：9.2 CUDNN：7.1.4 NCCL：2.2.13

以下是wheel文件的链接： wheel file

Answer 5

我已经安装了 Tensorflow 1.8 版。它推荐 CUDA 9.0。我正在使用具有 CUDA 计算能力 3.0 的 GTX 650M 卡，现在工作起来就像一个魅力。操作系统是 ubuntu 18.04。以下是详细步骤：

安装依赖项

我已经为我的 opencv 3.4 编译包含了 ffmpeg 和一些相关的包，如果不需要，请不要安装 运行以下命令：

sudo apt-get update 
sudo apt-get dist-upgrade -y
sudo apt-get autoremove -y
sudo apt-get upgrade
sudo add-apt-repository ppa:jonathonf/ffmpeg-3 -y
sudo apt-get update
sudo apt-get install build-essential -y
sudo apt-get install ffmpeg -y
sudo apt-get install cmake git libgtk2.0-dev pkg-config libavcodec-dev libavformat-dev libswscale-dev -y
sudo apt-get install python-dev libtbb2 libtbb-dev libjpeg-dev libpng-dev libtiff-dev libjasper-dev libdc1394-22-dev -y
sudo apt-get install libavcodec-dev libavformat-dev libswscale-dev libv4l-dev -y
sudo apt-get install libxvidcore-dev libx264-dev -y
sudo apt-get install unzip qtbase5-dev python-dev python3-dev python-numpy python3-numpy -y
sudo apt-get install libopencv-dev libgtk-3-dev libdc1394-22 libdc1394-22-dev libjpeg-dev libpng12-dev libtiff5-dev >libjasper-dev -y
sudo apt-get install libavcodec-dev libavformat-dev libswscale-dev libxine2-dev libgstreamer0.10-dev libgstreamer-plugins-base0.10-dev -y
sudo apt-get install libv4l-dev libtbb-dev libfaac-dev libmp3lame-dev libopencore-amrnb-dev libopencore-amrwb-dev libtheora-dev -y
sudo apt-get install libvorbis-dev libxvidcore-dev v4l-utils vtk6 -y
sudo apt-get install liblapacke-dev libopenblas-dev libgdal-dev checkinstall -y
sudo apt-get install libgtk-3-dev -y
sudo apt-get install libatlas-base-dev gfortran -y
sudo apt-get install qt-sdk -y
sudo apt-get install python2.7-dev python3.5-dev python-tk -y
sudo apt-get install cython libgflags-dev -y
sudo apt-get install tesseract-ocr -y
sudo apt-get install tesseract-ocr-eng -y 
sudo apt-get install tesseract-ocr-ell -y
sudo apt-get install gstreamer1.0-python3-plugin-loader -y
sudo apt-get install libdc1394-22-dev -y
sudo apt-get install openjdk-8-jdk
sudo apt-get install pkg-config zip g++-6 gcc-6 zlib1g-dev unzip  git
sudo wget https://bootstrap.pypa.io/get-pip.py
sudo python get-pip.py
sudo pip install -U pip
sudo pip install -U numpy
sudo pip install -U pandas
sudo pip install -U wheel
sudo pip install -U six

安装英伟达驱动

运行以下命令：

sudo add-apt-repository ppa:graphics-drivers/ppa
sudo apt-get update
sudo apt-get install nvidia-390 -y

重新启动并运行以下命令，它应该会为您提供如下图所示的详细信息：

gcc-6 和 g++-6 检查。

CUDA 9.0 需要 gcc-6 和 g++-6，运行以下命令：

cd /usr/bin 
sudo rm -rf gcc gcc-ar gcc-nm gcc-ranlib g++
sudo ln -s gcc-6 gcc
sudo ln -s gcc-ar-6 gcc-ar
sudo ln -s gcc-nm-6 gcc-nm
sudo ln -s gcc-ranlib-6 gcc-ranlib
sudo ln -s g++-6 g++

安装 CUDA 9.0

转到https://developer.nvidia.com/cuda-90-download-archive。选择选项：Linux->x86_64->Ubuntu->17.04->deb(local)。 下载主文件和两个补丁。 运行以下命令：

sudo dpkg -i cuda-repo-ubuntu1704-9-0-local_9.0.176-1_amd64.deb
sudo apt-key add /var/cuda-repo-9-0-local/7fa2af80.pub
sudo apt-get update
sudo apt-get install cuda

在您的PC上导航到第一个补丁并双击它，它将自动执行，第二个补丁也是如此。

在 ~/.bashrc 文件中添加以下行并重新启动它：

export PATH=/usr/local/cuda-9.0/bin${PATH:+:$PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda-9.0/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}

为 CUDA 9.0 安装 cudnn 7.1.4

从https://developer.nvidia.com/cudnn 下载 tar 文件并将其解压缩到您的下载文件夹 下载需要nvidia开发的登录，免费注册 运行以下命令：

cd ~/Downloads/cudnn-9.0-linux-x64-v7.1/cuda
sudo cp include/* /usr/local/cuda/include/
sudo cp lib64/libcudnn.so.7.1.4 lib64/libcudnn_static.a /usr/local/cuda/lib64/
cd /usr/lib/x86_64-linux-gnu
sudo ln -s libcudnn.so.7.1.4 libcudnn.so.7
sudo ln -s libcudnn.so.7 libcudnn.so

为 CUDA 9.0 安装 NCCL 2.2.12

从https://developer.nvidia.com/nccl 下载 tar 文件并将其解压缩到您的下载文件夹 下载需要nvidia开发的登录，免费注册 运行以下命令：

sudo mkdir -p /usr/local/cuda/nccl/lib /usr/local/cuda/nccl/include
cd ~/Downloads/nccl-repo-ubuntu1604-2.2.12-ga-cuda9.0_1-1_amd64/
sudo cp *.txt /usr/local/cuda/nccl
sudo cp include/*.h /usr/include/
sudo cp lib/libnccl.so.2.1.15 lib/libnccl_static.a /usr/lib/x86_64-linux-gnu/
sudo ln -s /usr/include/nccl.h /usr/local/cuda/nccl/include/nccl.h
cd /usr/lib/x86_64-linux-gnu
sudo ln -s libnccl.so.2.1.15 libnccl.so.2
sudo ln -s libnccl.so.2 libnccl.so
for i in libnccl*; do sudo ln -s /usr/lib/x86_64-linux-gnu/$i /usr/local/cuda/nccl/lib/$i; done

安装 Bazel（推荐手动安装 bazel 有效，参考：https://docs.bazel.build/versions/master/install-ubuntu.html#install-with-installer-ubuntu）

从https://github.com/bazelbuild/bazel/releases 下载“bazel-0.13.1-installer-darwin-x86_64.sh” 运行以下命令：

chmod +x bazel-0.13.1-installer-darwin-x86_64.sh
./bazel-0.13.1-installer-darwin-x86_64.sh --user
export PATH="$PATH:$HOME/bin"

编译张量流

我们将使用 CUDA 编译，使用 XLA JIT（哦，是的）和 jemalloc 作为 malloc 支持。所以我们为这些东西输入yes。 运行以下命令并按照运行配置的说明回答查询

git clone https://github.com/tensorflow/tensorflow 
git checkout r1.8
./configure
You have bazel 0.13.0 installed.
Please specify the location of python. [Default is /usr/bin/python]:
Please input the desired Python library path to use.  Default is [/usr/local/lib/python2.7/dist-packages]
Do you wish to build TensorFlow with jemalloc as malloc support? [Y/n]: y
jemalloc as malloc support will be enabled for TensorFlow.
Do you wish to build TensorFlow with Google Cloud Platform support? [Y/n]: n
No Google Cloud Platform support will be enabled for TensorFlow.
Do you wish to build TensorFlow with Hadoop File System support? [Y/n]: n
No Hadoop File System support will be enabled for TensorFlow.
Do you wish to build TensorFlow with Amazon S3 File System support? [Y/n]: n
No Amazon S3 File System support will be enabled for TensorFlow.
Do you wish to build TensorFlow with Apache Kafka Platform support? [Y/n]: n
No Apache Kafka Platform support will be enabled for TensorFlow.
Do you wish to build TensorFlow with XLA JIT support? [y/N]: y
XLA JIT support will be enabled for TensorFlow.
Do you wish to build TensorFlow with GDR support? [y/N]: n
No GDR support will be enabled for TensorFlow.
Do you wish to build TensorFlow with VERBS support? [y/N]: n
No VERBS support will be enabled for TensorFlow.
Do you wish to build TensorFlow with OpenCL SYCL support? [y/N]: n
No OpenCL SYCL support will be enabled for TensorFlow.
Do you wish to build TensorFlow with CUDA support? [y/N]: y
CUDA support will be enabled for TensorFlow.
Please specify the CUDA SDK version you want to use, e.g. 7.0. [Leave empty to default to CUDA 9.0]:
Please specify the location where CUDA 9.1 toolkit is installed. Refer to README.md for more details. [Default is /usr/local/cuda]:
Please specify the cuDNN version you want to use. [Leave empty to default to cuDNN 7.0]: 7.1.4
Please specify the location where cuDNN 7 library is installed. Refer to README.md for more details. [Default is /usr/local/cuda]:
Do you wish to build TensorFlow with TensorRT support? [y/N]: n
No TensorRT support will be enabled for TensorFlow.
Please specify the NCCL version you want to use. [Leave empty to default to NCCL 1.3]: 2.2.12
Please specify the location where NCCL 2 library is installed. Refer to README.md for more details. [Default is /usr/local/cuda]:/usr/local/cuda/nccl
Please specify a list of comma-separated Cuda compute capabilities you want to build with.
You can find the compute capability of your device at: https://developer.nvidia.com/cuda-gpus.
Please note that each additional compute capability significantly increases your build time and binary size. [Default is: 3.0]
Do you want to use clang as CUDA compiler? [y/N]: n
nvcc will be used as CUDA compiler.
Please specify which gcc should be used by nvcc as the host compiler. [Default is /usr/bin/x86_64-linux-gnu-gcc-7]: /usr/bin/gcc-6
Do you wish to build TensorFlow with MPI support? [y/N]: n
No MPI support will be enabled for TensorFlow.
Please specify optimization flags to use during compilation when bazel option "--config=opt" is specified [Default is -march=native]:
Would you like to interactively configure ./WORKSPACE for Android builds? [y/N]: n
Not configuring the WORKSPACE for Android builds.
Preconfigured Bazel build configs. You can use any of the below by adding "--config=<>" to your build command. See tools/bazel.rc for more details.
 --config=mkl          # Build with MKL support.

 --config=monolithic   # Config for mostly static monolithic build.

Configuration finished

现在要编译 tensorflow，运行下面的命令，这非常消耗 RAM 并且需要时间。如果您有大量 RAM，则可以从下面的行中删除“--local_resources 2048,.5,1.0”，否则这将适用于 2 GB 的 RAM

bazel build --config=opt --config=cuda --local_resources 2048,.5,1.0 //tensorflow/tools/pip_package:build_pip_package

编译完成后，您将看到如下图所示的内容，确认编译成功

构建wheel文件，运行如下：

bazel-bin/tensorflow/tools/pip_package/build_pip_package /tmp/tensorflow_pkg

使用pip安装生成的wheel文件

sudo pip install /tmp/tensorflow_pkg/tensorflow*.whl

现在要在设备上进行探索，您可以运行 tensorflow，下图是 ipython 终端上的展示