Home | History | Annotate | only in /external/tensorflow/tensorflow/tools/ci_build
Up to higher level directory
NameDateSize
builds/22-Oct-2020
ci_build.sh22-Oct-20206.6K
ci_parameterized_build.sh22-Oct-202023.1K
ci_rbe_docker_build.sh22-Oct-20202.5K
ci_sanity.sh22-Oct-202019.4K
code_link_check.sh22-Oct-20201.3K
copy_binary.py22-Oct-20204K
Dockerfile.android22-Oct-20202.6K
Dockerfile.cmake22-Oct-20201.3K
Dockerfile.cpu22-Oct-2020611
Dockerfile.cpu.mpi22-Oct-2020734
Dockerfile.cpu.ppc64le22-Oct-2020665
Dockerfile.custom_op22-Oct-2020612
Dockerfile.custom_op_gpu22-Oct-2020661
Dockerfile.debian.jessie.cpu22-Oct-2020969
Dockerfile.gpu22-Oct-20201.5K
Dockerfile.gpu.ppc64le22-Oct-20201.1K
Dockerfile.hadoop22-Oct-2020579
Dockerfile.pi22-Oct-2020806
Dockerfile.pi-python322-Oct-2020814
Dockerfile.rbe.cpu22-Oct-2020575
Dockerfile.rbe.cuda10.0-cudnn7-ubuntu14.0422-Oct-20203.9K
Dockerfile.rbe.cuda9.0-cudnn7-ubuntu14.0422-Oct-20204.1K
Dockerfile.rbe.gpu22-Oct-20201K
Dockerfile.rocm22-Oct-20202.5K
gpu_build/22-Oct-2020
install/22-Oct-2020
linux/22-Oct-2020
osx/22-Oct-2020
pep822-Oct-2020283
pi/22-Oct-2020
protobuf/22-Oct-2020
pylintrc22-Oct-202010.7K
README.md22-Oct-20203.1K
remote/22-Oct-2020
update_version.py22-Oct-202011.3K
windows/22-Oct-2020
xla/22-Oct-2020

README.md

      1 # TensorFlow Builds
      2 
      3 This directory contains all the files and setup instructions to run all the
      4 important builds and tests. You can run it yourself!
      5 
      6 ## Run It Yourself
      7 
      8 You have two options when running TensorFlow tests locally on your
      9 machine. First, using docker, you can run our Continuous Integration
     10 (CI) scripts on tensorflow devel images. The other option is to install
     11 all TensorFlow dependencies on your machine and run the scripts
     12 natively on your system.
     13 
     14 ### Run TensorFlow CI Scripts using Docker
     15 
     16 1.  Install Docker following the [instructions on the docker website](https://docs.docker.com/engine/installation/).
     17 
     18 2.  Start a container with one of the devel images here:
     19     https://hub.docker.com/r/tensorflow/tensorflow/tags/.
     20 
     21 3.  Based on your choice of the image, pick one of the scripts under
     22     https://github.com/tensorflow/tensorflow/tree/master/tensorflow/tools/ci_build/linux
     23     and run them from the TensorFlow repository root.
     24 
     25 ### Run TensorFlow CI Scripts Natively on your Machine
     26 
     27 1.  Follow the instructions at https://www.tensorflow.org/install/source,
     28     but stop when you get to the section "Configure the installation". You do not
     29     need to configure the installation to run the CI scripts.
     30 
     31 2.  Pick the appropriate OS and python version you have installed,
     32     and run the script under tensorflow/tools/ci_build/<OS>.
     33 
     34 ## TensorFlow Continuous Integration
     35 
     36 To verify that new changes dont break TensorFlow, we run builds and
     37 tests on either [Jenkins](https://jenkins-ci.org/) or a CI system
     38 internal to Google.
     39 
     40 We can trigger builds and tests on updates to master or on each pull
     41 request. Contact one of the repository maintainers to trigger builds
     42 on your pull request.
     43 
     44 ### View CI Results
     45 
     46 The Pull Request will show if the change passed or failed the checks.
     47 
     48 From the pull request, click **Show all checks** to see the list of builds
     49 and tests. Click on **Details** to see the results from Jenkins or the internal
     50 CI system.
     51 
     52 Results from Jenkins are displayed in the Jenkins UI. For more information,
     53 see the [Jenkins documentation](https://jenkins.io/doc/).
     54 
     55 Results from the internal CI system are displayed in the Build Status UI. In
     56 this UI, to see the logs for a failed build:
     57 
     58 *   Click on the **INVOCATION LOG** tab to see the invocation log.
     59 
     60 *   Click on the **ARTIFACTS** tab to see a list of all artifacts, including logs.
     61 
     62 *   Individual test logs may be available. To see these logs, from the **TARGETS**
     63     tab, click on the failed target. Then, click on the **TARGET LOG** tab to see
     64     its test log.
     65 
     66     If youre looking at target that is sharded or a test that is flaky, then
     67     the build tool divided the target into multiple shards or ran the test
     68     multiple times. Each test log is specific to the shard, run, and attempt.
     69     To see a specific log:
     70 
     71     1.  Click on the log icon that is on the right next to the shard, run,
     72         and attempt number.
     73 
     74     2.  In the grid that appears on the right, click on the specific shard,
     75         run, and attempt to view its log. You can also type the desired shard,
     76         run, or attempt number in the field above its grid.
     77