我从源(文档)安装tensorflow .
Cuda驱动版:
nvcc: NVIDIA (R) Cuda compiler driver
Cuda compilation tools, release 7.5, V7.5.17
Run Code Online (Sandbox Code Playgroud)
当我运行以下命令时:
bazel-bin/tensorflow/cc/tutorials_example_trainer --use_gpu
Run Code Online (Sandbox Code Playgroud)
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 …Run Code Online (Sandbox Code Playgroud) 当我bazel info --local_cpu_resources HOST_CPUS继续macos奔跑时ubuntu
`bazel-bin: .../9dcdc19e81f948a3daf2cc314e0d4bf1/execroot/demo/bazel-out/darwin-fastbuild/bin`
`bazel-genfiles: .../9dcdc19e81f948a3daf2cc314e0d4bf1/execroot/demo/bazel-out/darwin-fastbuild/bin`
`bazel-testlogs: .../9dcdc19e81f948a3daf2cc314e0d4bf1/execroot/demo/bazel-out/darwin-fastbuild/testlogs`
Run Code Online (Sandbox Code Playgroud)
`bazel-bin: /root/.cache/bazel/_bazel_root/d89113122c03ccb06b1d63e7c96f20e4/execroot/demo/bazel-out/k8-fastbuild/bin`
`bazel-genfiles: /root/.cache/bazel/_bazel_root/d89113122c03ccb06b1d63e7c96f20e4/execroot/demo/bazel-out/k8-fastbuild/bin`
`bazel-testlogs: /root/.cache/bazel/_bazel_root/d89113122c03ccb06b1d63e7c96f20e4/execroot/demo/bazel-out/k8-fastbuild/testlogs`
Run Code Online (Sandbox Code Playgroud)
Darwin可能意味着OS在macos.k8意思ubuntu?windows,会输出什么?为什么?https://docs.bazel.build/versions/master/be/make-variables.html#predefined_variables
https://docs.bazel.build/versions/master/guide.html#distinct_host_configuration-true-default
root@608a654a26de:/home/servlet/demo# bazel info --local_cpu_resources HOST_CPUS
Extracting Bazel installation...
Starting local Bazel server and connecting to it...
bazel-bin: /root/.cache/bazel/_bazel_root/d89113122c03ccb06b1d63e7c96f20e4/execroot/demo/bazel-out/k8-fastbuild/bin
bazel-genfiles: /root/.cache/bazel/_bazel_root/d89113122c03ccb06b1d63e7c96f20e4/execroot/demo/bazel-out/k8-fastbuild/bin
bazel-testlogs: /root/.cache/bazel/_bazel_root/d89113122c03ccb06b1d63e7c96f20e4/execroot/demo/bazel-out/k8-fastbuild/testlogs
character-encoding: file.encoding = ISO-8859-1, defaultCharset = ISO-8859-1
command_log: /root/.cache/bazel/_bazel_root/d89113122c03ccb06b1d63e7c96f20e4/command.log
committed-heap-size: 40MB
execution_root: /root/.cache/bazel/_bazel_root/d89113122c03ccb06b1d63e7c96f20e4/execroot/demo …Run Code Online (Sandbox Code Playgroud) 有没有办法告诉Bazel什么时候可以使用多少CPU内核?
我在VMware Workstation上构建TensorFlow并且作为虚拟机我可以调整虚拟机的处理器和核心数量.
在构建TensorFlow的过程中,我发现只使用一个核心工作.当我给Workstation提供四个内核并构建TensorFlow时,它最终将系统停止到我必须重启的程度.
如果我等了几个小时(不管它一夜之间),它有时会返回以下错误:
gcc:内部编译器错误:已杀死(程序cc1plus)
虽然我可以使用虚拟机配置选项更改内核数量,但我更愿意这样做,而无需关闭并重新启动虚拟机.
In Bazel, given a build target, how would a script (which is running outside of Bazel) get the path to the generated file?
Scenario: I'm using Bazel to do the build, and then when it's done, I want to copy the result to a server. I just need to know what files to copy. I could hard-code the list of files, but I would prefer not to do that.
A simple example: This Bazel script:
genrule(
name = "main",
srcs …Run Code Online (Sandbox Code Playgroud) 既然Bazel(http://bazel.io/)已经开源了,那么我是否可以通过增量流程逐步迁移(一个大型存储库)从Maven到Bazel?
使用Bazel构建使用OpenCV库的C++代码的最佳方法是什么?即,BUILD规则是什么样的?
Bazel.io有外部依赖的文档,但不是很清楚.
我想将ngx-socket-io添加到我的 Angular 应用程序中。我使用 Bazel 来运行我的 Angular 开发服务器。不幸的是ngx-socket-io,ts_devserver开箱即用似乎并不适用。我在浏览器控制台中收到此错误:
Uncaught TypeError: XMLHttpRequest is not a constructor
at ts_scripts.js?v=1587802098203:16776
at Object.23.../transport (ts_scripts.js?v=1587802098203:16780)
at o (ts_scripts.js?v=1587802098203:11783)
Run Code Online (Sandbox Code Playgroud)
它似乎是由xmlhttprequest-ssl引起的,它是engine.io-client 的一个依赖项,它是ngx-socket-io. 但是这个问题只发生在ts_devserver 上。在生产中运行 Angular 应用程序完全正常。
你可以自己轻松尝试:https : //github.com/flolu/bazel-socketio-issue
只需运行yarn install然后yarn dev(它会导致浏览器控制台 @ http://localhost:4200 中的错误)。并注意yarn prod@ http://localhost:8080工作正常!
目前,Windows 上似乎还有另一个问题。因此,如果您运行的是 Mac 或 Linux,则只能尝试示例存储库
java测试规则上的bazel testvs有什么区别?bazel run当我运行该bazel run命令时,我的测试顺利通过,但bazel test出现错误。如果需要,我可以提供更多详细信息,但我对这两个命令之间的一般差异感兴趣。
或者更好的是,如何安装 bazel。我一直使用 cd 进入桌面上的文件夹,然后我使用git clone https://github.com/bazelbuild/bazelisk然后?我现在应该做什么,老实说,我没有找到任何真正的说明,我现在有很多文件夹,我现在应该运行像 bazelisk.py 这样的特殊文件吗?或者喜欢建造它?我是否应该将 github 存储库克隆到一个特殊位置?如前所述,我只是在桌面上选择一个文件夹,但我添加了 Path 环境变量的路径,我添加了 ,C:\Users\user\Desktop\folder\bazel因为C:\Users\user\Desktop\folder\bazel\bazelisk我不确定应该走多远。因此,如果有人能告诉我如何安装 bazel,我将非常感激,我只需要它用于 python 项目,我使用的是 python 3.6.2,并且使用的是 windows 10。我不知道现在该怎么办,因为我读过的安装步骤只说到 git clone 存储库,但是当我输入 cmd 并键入bazel或bazel helpcmd 时找不到该命令。
Bazel在清理编译(java)期间显示所有编译警告.
但是 - 当我们重新运行bazel build时 - 我们失去了所有的警告.
是否有任何标志可以显示缓存目标的原始警告?
示例回购: https ://github.com/or-shachar/bazel-sample-repo/tree/warning(branch warning)
首次输出:
? bazel build //...
INFO: Found 1 target...
INFO: From Building src/main/java/com/example/libmy_warning_lib.jar (1 source file):
src/main/java/com/example/MyLib.java:12: warning: [static] static variable should be qualified by type name, MyLib, instead of by an expression
this.x = this.x + 2.1;
^
src/main/java/com/example/MyLib.java:12: warning: [static] static variable should be qualified by type name, MyLib, instead of by an expression
this.x = this.x + 2.1;
^
Target //src/main/java/com/example:my_warning_lib up-to-date: …Run Code Online (Sandbox Code Playgroud) bazel ×10
build ×2
python ×2
angular ×1
compilation ×1
cpu-cores ×1
git ×1
gpu ×1
h2 ×1
installation ×1
java ×1
junit4 ×1
maven ×1
migration ×1
opencv ×1
python-3.x ×1
socket.io ×1
spring ×1
tensorflow ×1
testing ×1
typescript ×1
ubuntu ×1