Ubuntu 16.04 + CUDA toolkit 10.1 + cudNN7.6 + bazel 0.26.1

保證cuDNN跟CUDA SDK的版本一致即可.

從源碼編譯的話需要查看顯卡的計算能力,可在查看GPU計算能力，在配置的時候填入即可，跟CUDA SDK的版本無關(guān)

▲必須注冊賬號后，才能選擇下載

▲記住上述幾個要點， 以及最好先看下文章最后我的踩坑記錄，確保自己知道可能會有這些坑

▲.全程保持能連接外網(wǎng)（翻墻）

安裝NVIDIA驅(qū)動

<pre spellcheck="false" class="md-fences md-end-block md-fences-with-lineno ty-contain-cm modeLoaded" lang="shell" cid="n10" mdtype="fences" style="box-sizing: border-box; overflow: visible; font-family: Monaco, Consolas, "Andale Mono", "DejaVu Sans Mono", monospace; margin-top: 0px; margin-bottom: 20px; font-size: 0.9rem; display: block; break-inside: avoid; text-align: left; white-space: normal; background: rgb(51, 51, 51); position: relative !important; padding: 10px 10px 10px 0px; width: inherit; color: rgb(184, 191, 198); font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;"> $ ubuntu-drivers devices # 查看

查看NVIDIA驅(qū)動版本

$ sudo dpkg --list | grep nvidia-*</pre>

顯卡驅(qū)動與CUDA版本對應(yīng)關(guān)系

最新的根據(jù)官網(wǎng)查看

image

下載NVIDIA驅(qū)動

▲注意:得對應(yīng)匹配!!!

安裝完成后，可以通過nvidia-smi查看GPU使用情況

CUDA toolkit 10.1

安裝之前先看顯卡的驅(qū)動版本

<pre spellcheck="false" class="md-fences md-end-block md-fences-with-lineno ty-contain-cm modeLoaded" lang="shell" cid="n20" mdtype="fences" style="box-sizing: border-box; overflow: visible; font-family: Monaco, Consolas, "Andale Mono", "DejaVu Sans Mono", monospace; margin-top: 0px; margin-bottom: 20px; font-size: 0.9rem; display: block; break-inside: avoid; text-align: left; white-space: normal; background: rgb(51, 51, 51); position: relative !important; padding: 10px 10px 10px 0px; width: inherit; color: rgb(184, 191, 198); font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;"> # 安裝過程
sudo apt update
$ sudo apt install cuda</pre>

設(shè)置環(huán)境變量

<pre spellcheck="false" class="md-fences md-end-block md-fences-with-lineno ty-contain-cm modeLoaded" lang="shell" cid="n22" mdtype="fences" style="box-sizing: border-box; overflow: visible; font-family: Monaco, Consolas, "Andale Mono", "DejaVu Sans Mono", monospace; margin-top: 0px; margin-bottom: 20px; font-size: 0.9rem; display: block; break-inside: avoid; text-align: left; white-space: normal; background: rgb(51, 51, 51); position: relative !important; padding: 10px 10px 10px 0px; width: inherit; color: rgb(184, 191, 198); font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;"> $ sudo vim ~/.bashrc

設(shè)置環(huán)境變量

PATH
LD_LIBRARY_PATH

生效

$ source ~/.bashrc</pre>

驗證安裝完成

<pre spellcheck="false" class="md-fences md-end-block md-fences-with-lineno ty-contain-cm modeLoaded" lang="shell" cid="n24" mdtype="fences" style="box-sizing: border-box; overflow: visible; font-family: Monaco, Consolas, "Andale Mono", "DejaVu Sans Mono", monospace; margin-top: 0px; margin-bottom: 20px; font-size: 0.9rem; display: block; break-inside: avoid; text-align: left; white-space: normal; background: rgb(51, 51, 51); position: relative !important; padding: 10px 10px 10px 0px; width: inherit; color: rgb(184, 191, 198); font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;"> cd /usr/local/cuda/samples/1_Utilities/deviceQuery
./deviceQuery</pre>

△.如果顯示Result:PASS才是安裝成功。這一步一定要保證錯誤，不然之后bazel編譯時報錯就很難分析問題！?。。?！(只有在驅(qū)動安裝好的情況下才能運行成功,不然會報no CUDA-capable device is detected，或者GPU is lost)

如果失敗,卸載重裝

<pre spellcheck="false" class="md-fences md-end-block md-fences-with-lineno ty-contain-cm modeLoaded" lang="shell" cid="n27" mdtype="fences" style="box-sizing: border-box; overflow: visible; font-family: Monaco, Consolas, "Andale Mono", "DejaVu Sans Mono", monospace; margin-top: 0px; margin-bottom: 20px; font-size: 0.9rem; display: block; break-inside: avoid; text-align: left; white-space: normal; background: rgb(51, 51, 51); position: relative !important; padding: 10px 10px 10px 0px; width: inherit; color: rgb(184, 191, 198); font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;"> sudo ./uninstall_cuda_9.2.pl</pre>

安裝 Python 和 TensorFlow 軟件包依賴項

<pre spellcheck="false" class="md-fences md-end-block md-fences-with-lineno ty-contain-cm modeLoaded" lang="shell" cid="n30" mdtype="fences" style="box-sizing: border-box; overflow: visible; font-family: Monaco, Consolas, "Andale Mono", "DejaVu Sans Mono", monospace; margin-top: 0px; margin-bottom: 20px; font-size: 0.9rem; display: block; break-inside: avoid; text-align: left; white-space: normal; background: rgb(51, 51, 51); position: relative !important; padding: 10px 10px 10px 0px; width: inherit; color: rgb(184, 191, 198); font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;"> sudo apt install python-dev python-pip # or python3-dev python3-pip</pre>

安裝 TensorFlow pip 軟件包依賴項（如果使用虛擬環(huán)境，請省略 --user 參數(shù)）：

<pre spellcheck="false" class="md-fences md-end-block md-fences-with-lineno ty-contain-cm modeLoaded" lang="bash" cid="n32" mdtype="fences" style="box-sizing: border-box; overflow: visible; font-family: Monaco, Consolas, "Andale Mono", "DejaVu Sans Mono", monospace; margin-top: 0px; margin-bottom: 20px; font-size: 0.9rem; display: block; break-inside: avoid; text-align: left; white-space: normal; background: rgb(51, 51, 51); position: relative !important; padding: 10px 10px 10px 0px; width: inherit; color: rgb(184, 191, 198); font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;"> pip install -U --user pip six numpy wheel setuptools mock future>=0.17.1
pip install -U --user keras_applications==1.0.6 --no-deps
pip install -U --user keras_preprocessing==1.0.5 --no-deps</pre>

cudNN 7.6.3

選擇cuDNN Library for Linux

<pre spellcheck="false" class="md-fences md-end-block md-fences-with-lineno ty-contain-cm modeLoaded" lang="shell" cid="n35" mdtype="fences" style="box-sizing: border-box; overflow: visible; font-family: Monaco, Consolas, "Andale Mono", "DejaVu Sans Mono", monospace; margin-top: 0px; margin-bottom: 20px; font-size: 0.9rem; display: block; break-inside: avoid; text-align: left; white-space: normal; background: rgb(51, 51, 51); position: relative !important; padding: 10px 10px 10px 0px; width: inherit; color: rgb(184, 191, 198); font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;"> # 解壓cudNN
cd /usr/local/cuda/
$ tar -xvf cudnn-8.0-linux-x64-v5.1.tgz
?
sudo cp cuda/include/cudnn.h /usr/local/cuda/include
sudo cp cuda/lib64/libcudnn* /usr/local/cuda/lib64
sudo chmod a+r /usr/local/cuda/include/cudnn.h /usr/local/cuda/lib64/libcudnn*</pre>

Bazel

構(gòu)建Tensorflow需要用bazel,而wget下載很慢,需要代理,可以試著去github上下載后再傳到服務(wù)器

后來./configure配置TensorFlow時,提示Bazel版本太高...(這個問題出現(xiàn)在我下載的不是github下載下來的tensorflow，網(wǎng)盤下載的可能版本太低了)

<pre spellcheck="false" class="md-fences md-end-block md-fences-with-lineno ty-contain-cm modeLoaded" lang="shell" cid="n40" mdtype="fences" style="box-sizing: border-box; overflow: visible; font-family: Monaco, Consolas, "Andale Mono", "DejaVu Sans Mono", monospace; margin-top: 0px; margin-bottom: 20px; font-size: 0.9rem; display: block; break-inside: avoid; text-align: left; white-space: normal; background: rgb(51, 51, 51); position: relative !important; padding: 10px 10px 10px 0px; width: inherit; color: rgb(184, 191, 198); font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;"> WARNING: --batch mode is deprecated. Please instead explicitly shut down your Bazel server using the command "bazel shutdown".
You have bazel 0.29.1 installed.
Please downgrade your bazel installation to version 0.26.1 or lower to build TensorFlow! To downgrade: download the installer for the old version (from https://github.com/bazelbuild/bazel/releases) then run the installer.</pre>

從源碼構(gòu)建和安裝Tensorflow

第一次安裝的是2.0的版本

<pre spellcheck="false" class="md-fences md-end-block md-fences-with-lineno ty-contain-cm modeLoaded" lang="bash" cid="n44" mdtype="fences" style="box-sizing: border-box; overflow: visible; font-family: Monaco, Consolas, "Andale Mono", "DejaVu Sans Mono", monospace; margin-top: 0px; margin-bottom: 20px; font-size: 0.9rem; display: block; break-inside: avoid; text-align: left; white-space: normal; background: rgb(51, 51, 51); position: relative !important; padding: 10px 10px 10px 0px; width: inherit; color: rgb(184, 191, 198); font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;"> git clone https://github.com/tensorflow/tensorflow.git
cd tensorflow
(tf) apollo3d@apollo1:~/Downloads/tensorflow-master$ ./configure
WARNING: --batch mode is deprecated. Please instead explicitly shut down your Bazel server using the command "bazel shutdown".
You have bazel 0.26.0 installed.
Please specify the location of python. [Default is /home/apollo3d/tf/bin/python]: /usr/bin/python3
?
?
Found possible Python library paths:
/usr/local/lib/python3.5/dist-packages
/usr/lib/python3/dist-packages
Please input the desired Python library path to use. Default is [/usr/local/lib/python3.5/dist-packages]
?
Do you wish to build TensorFlow with XLA JIT support? [Y/n]:
XLA JIT support will be enabled for TensorFlow.
?
Do you wish to build TensorFlow with OpenCL SYCL support? [y/N]:
No OpenCL SYCL support will be enabled for TensorFlow.
?
Do you wish to build TensorFlow with ROCm support? [y/N]:
No ROCm 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.
?
Do you wish to build TensorFlow with TensorRT support? [y/N]:
No TensorRT support will be enabled for TensorFlow.
?
Found CUDA 10.1 in:
/usr/local/cuda/lib64
/usr/local/cuda/include
Found cuDNN 7 in:
/usr/local/cuda/lib64
/usr/local/cuda/include
?
?
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, and that TensorFlow only supports compute capabilities >= 3.5 [Default is: 3.5,7.0]: 5.0
?
?
Do you want to use clang as CUDA compiler? [y/N]: N
Clang will be used as CUDA compiler.
?
Do you wish to download a fresh release of clang? (Experimental) [y/N]: N
Clang will be downloaded and used to compile tensorflow.
?
Please specify optimization flags to use during compilation when bazel option "--config=opt" is specified [Default is -march=native -Wno-sign-compare]: --config=v2
?
?
Would you like to interactively configure ./WORKSPACE for Android builds? [y/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 .bazelrc for more details.
--config=mkl # Build with MKL support.
--config=monolithic # Config for mostly static monolithic build.
--config=ngraph # Build with Intel nGraph support.
--config=numa # Build with NUMA support.
--config=dynamic_kernels # (Experimental) Build kernels into separate shared objects.
--config=v2 # Build TensorFlow 2.x instead of 1.x.
Preconfigured Bazel build configs to DISABLE default on features:
--config=noaws # Disable AWS S3 filesystem support.
--config=nogcp # Disable GCP support.
--config=nohdfs # Disable HDFS support.
--config=nonccl # Disable NVIDIA NCCL support.
Configuration finished
</pre>

第二次1.14

<pre spellcheck="false" class="md-fences md-end-block md-fences-with-lineno ty-contain-cm modeLoaded" lang="bash" cid="n47" mdtype="fences" style="box-sizing: border-box; overflow: visible; font-family: Monaco, Consolas, "Andale Mono", "DejaVu Sans Mono", monospace; margin-top: 0px; margin-bottom: 20px; font-size: 0.9rem; display: block; break-inside: avoid; text-align: left; white-space: normal; background: rgb(51, 51, 51); position: relative !important; padding: 10px 10px 10px 0px; width: inherit; color: rgb(184, 191, 198); font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;"> (pythonEnv) apollo3d@apollo1:~/Downloads/tensorflow$ ./configure
WARNING: Running Bazel server needs to be killed, because the startup options are different.
WARNING: --batch mode is deprecated. Please instead explicitly shut down your Bazel server using the command "bazel shutdown".
You have bazel 0.26.0 installed.
Please specify the location of python. [Default is /home/apollo3d/pythonEnv/bin/python]:
?
?
Traceback (most recent call last):
File "<string>", line 1, in <module>
AttributeError: module 'site' has no attribute 'getsitepackages'
Found possible Python library paths:
/home/apollo3d/pythonEnv/lib/python3.5/site-packages
Please input the desired Python library path to use. Default is [/home/apollo3d/pythonEnv/lib/python3.5/site-packages]
?
Do you wish to build TensorFlow with XLA JIT support? [Y/n]:
XLA JIT 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 ROCm support? [y/N]:
No ROCm 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.
?
Do you wish to build TensorFlow with TensorRT support? [y/N]:
No TensorRT support will be enabled for TensorFlow.
?
Found CUDA 10.1 in:
/usr/local/cuda/lib64
/usr/local/cuda/include
Found cuDNN 7 in:
/usr/local/cuda/lib64
/usr/local/cuda/include
?
?
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, and that TensorFlow only supports compute capabilities >= 3.5 [Default is: 5.2]: 5.0
?
?
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/gcc]:
?
?
Do you wish to build TensorFlow with MPI support? [y/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 -Wno-sign-compare]:
?
?
Would you like to interactively configure ./WORKSPACE for Android builds? [y/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 .bazelrc for more details.
--config=mkl # Build with MKL support.
--config=monolithic # Config for mostly static monolithic build.
--config=gdr # Build with GDR support.
--config=verbs # Build with libverbs support.
--config=ngraph # Build with Intel nGraph support.
--config=numa # Build with NUMA support.
--config=dynamic_kernels # (Experimental) Build kernels into separate shared objects.
Preconfigured Bazel build configs to DISABLE default on features:
--config=noaws # Disable AWS S3 filesystem support.
--config=nogcp # Disable GCP support.
--config=nohdfs # Disable HDFS support.
--config=noignite # Disable Apache Ignite support.
--config=nokafka # Disable Apache Kafka support.
--config=nonccl # Disable NVIDIA NCCL support.
Configuration finished
?
...
?
INFO: Elapsed time: 4826.834s, Critical Path: 298.31s
INFO: 24978 processes: 24978 local.
INFO: Build completed successfully, 26636 total actions

大概用了一個半小時</pre>

△.期間可能多次出現(xiàn)ERROR:[GET returned 404 Not Found, connect timed out],繼續(xù)重試即可

編譯 pip 軟件包

<pre spellcheck="false" class="md-fences md-end-block md-fences-with-lineno ty-contain-cm modeLoaded" lang="bash" cid="n51" mdtype="fences" style="box-sizing: border-box; overflow: visible; font-family: Monaco, Consolas, "Andale Mono", "DejaVu Sans Mono", monospace; margin-top: 0px; margin-bottom: 20px; font-size: 0.9rem; display: block; break-inside: avoid; text-align: left; white-space: normal; background: rgb(51, 51, 51); position: relative !important; padding: 10px 10px 10px 0px; width: inherit; color: rgb(184, 191, 198); font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial;"> bazel build -c --config=opt --config=cuda //tensorflow/tools/pip_package:build_pip_package

-c 不鏈接直接編譯

Starting local Bazel server and connecting to it...
WARNING: The following configs were expanded more than once: [cuda_clang, using_cuda, download_clang_use_lld]. For repeatable flags, repeats are counted twice and may lead to unexpected behavior.
WARNING: option '--crosstool_top' was expanded to from both option '--config=cuda_clang' (source /home/apollo3d/Downloads/tensorflow-master/.tf_configure.bazelrc) and option '--config=download_clang' (source /home/apollo3d/Downloads/tensorflow-master/.tf_configure.bazelrc)
?
$ bazel-bin/tensorfLow/tools/pip_package/build_pip_package ~/
tensorflow/bin # 傳入一個表示Python whell文件存儲路徑的參數(shù)
?

在虛擬環(huán)境打開的情況下安裝</pre>

附錄：

windows上安裝cuda時一直顯示不能安裝，就是系統(tǒng)默認(rèn)安裝的版本太低了，或者根本就沒有安裝。只有到顯卡驅(qū)動下載安裝了驅(qū)動后，才能正常安裝CUDA

▲正確安裝驅(qū)動真的很重要?。。?/strong>